mam-e17 rat model impairments on a novel continuous ... · original investigation mam-e17 rat model...

ORIGINAL INVESTIGATION

MAM-E17 rat model impairments on a novel continuousperformance task: effects of potential cognitive enhancing drugs

Adam C. Mar1,2,3,4 & Simon R. O. Nilsson1,2,3,4& Begoña Gamallo-Lana1,2,3,4 &

Ming Lei3,4,5 & Theda Dourado3,4 & Johan Alsiö3,4,6 & Lisa M. Saksida3,4,7,8,9 &

Timothy J. Bussey3,4,7,8,9 & Trevor W. Robbins3,4

Received: 29 March 2017 /Accepted: 18 June 2017 /Published online: 26 July 2017# The Author(s) 2017. This article is an open access publication

AbstractRationale Impairments in attention and inhibitory control areendophenotypicmarkers of neuropsychiatric disorders such asschizophrenia and represent key targets for therapeutic man-agement. Robust preclinical models and assays sensitive toclinically relevant treatments are crucial for improving cogni-tive enhancement strategies.Objectives We assessed a rodent model with neural and be-havioral features relevant to schizophrenia (gestational day 17methylazoxymethanol acetate treatment (MAM-E17)) on anovel test of attention and executive function, and examinedthe impact of putative nootropic drugs.Methods MAM-E17 and sham control rats were trained on anovel touchscreen-based rodent continuous performance test(rCPT) designed to closely mimic the human CPT paradigm.Performance following acute, systemic treatment with an arrayof pharmacological compounds was investigated.Results Two cohorts of MAM-E17 rats were impaired onrCPT performance including deficits in sensitivity (d′) and

increased false alarm rates (FARs). Sulpiride (0–30 mg/kg)dose-dependently reduced elevated FAR in MAM-E17 ratswhereas low-dose modafinil (8 mg/kg) only improved d′ insham controls. ABT-594 (5.9–19.4 μg/kg) and modafinil(64 mg/kg) showed expected stimulant-like effects, whileLSN2463359 (5 mg/kg), RO493858 (10 mg/kg), atomoxetine(0.3–1 mg/kg), and sulpiride (30 mg/kg) showed expectedsuppressant effects on performance across all animals.Donepezil (0.1–1 mg/kg) showed near-significant enhance-ments in d′, and EVP-6124 (0.3–3 mg/kg) exerted no effectsin the rCPT paradigm.Conclusion The MAM-E17 model exhibits robust and repli-cable impairments in rCPT performance that resemble atten-tion and inhibitory control deficits seen in schizophrenia.Pharmacological profiles were highly consistent with knowndrug effects on cognition in preclinical and clinical studies.The rCPT is a sensitive and reliable tool with high translation-al potential for understanding the etiology and treatment ofdisorders affecting attention and executive dysfunction.

Electronic supplementary material The online version of this article(doi:10.1007/s00213-017-4679-5) contains supplementary material,which is available to authorized users.

* Adam C. [email protected]

1 Neuroscience Institute, New York University Medical Center, NewYork, NY 10016, USA

2 Department of Neuroscience and Physiology, New York UniversityMedical Center, New York, NY, USA

3 Department of Psychology, University of Cambridge,Cambridge, UK

4 MRC and Wellcome Trust Behavioural and Clinical NeuroscienceInstitute, University of Cambridge, Cambridge, UK

5 Department of Health Industry Management, Beijing InternationalStudies University, 1 Dingfuzhuang Nanli, Beijing, China

6 Department of Neuroscience, Unit of Functional Neurobiology,University of Uppsala, Uppsala, Sweden

7 Molecular Medicine Research Group, Robarts Research Institute,Western University, London, ON, Canada

8 Department of Physiology and Pharmacology, Schulich School ofMedicine and Dentistry, Western University, London, ON, Canada

9 The Brain and Mind Institute, Western University, London, ON,Canada

Psychopharmacology (2017) 234:2837–2857DOI 10.1007/s00213-017-4679-5

http://dx.doi.org/10.1007/s00213-017-4679-5

mailto:[email protected]

http://crossmark.crossref.org/dialog/?doi=10.1007/s00213-017-4679-5&domain=pdf

Keywords Attention .MAM-E17 . Executive function .

Touch screen . Behavioral pharmacology . Schizophrenia .

ADHD . Animal model

Introduction

Neurological and psychiatric disorders such as schizophre-nia, attention deficit hyperactivity disorder and Alzheimer’sdisease have profiles of cognitive impairment that are criticaltargets for therapeutic remediation. Central to these profilesare deficits in executive function including impairments insustained goal-directed action, attentional and inhibitorycontrol and regulation of processing speed (Buchanan et al.1997; Bilder et al. 2000; Knowles et al. 2010). The estab-lishment of robust preclinical models that appropriatelytranslate prominent cognitive features of a disorder, alongwith the development of valid behavioral assays that aresensitive to model impairments as well as clinically relevanttreatments, is an important goal for improving cognitive en-hancement strategies.

A widely used model of neurodevelopmental hallmarksof schizophrenia involves administration of the mitoticneurotoxin methylazoxymethanol acetate (MAM) topregnant rat dams at gestational day 17. The offspring ofMAM-treated dams (gestational day 17 methylazoxy-methanol acetate treatment, MAM-E17) show pronouncedneurodevelopmental alterations within limbic and corticalbrain structures (Moore et al. 2006). Histological and electro-physiological evidence reveals structural and functional dis-ruptions predominantly within hippocampal and frontostriatalcircuitries in the MAM-E17 model (Gourevitch et al. 2004;Penschuck et al. 2006; Moore et al. 2006; Lodge and Grace2008;Matricon et al. 2010; Hradetzky et al. 2012; Gastambideet al. 2012; Phillips et al. 2012; Snyder et al. 2012). Thesedisruptions mirror well-established abnormalities in individ-uals afflicted with schizophrenia: hippocampal and corticaldisorganization including aberrant expression of markers forinhibitory (Benes et al. 1996; Lewis 2000) and excitatory(Deakin et al. 1989; Tsai et al. 1995) signaling efficacy andstriatal dopamine (DA) hyperfunction (Abi-Dargham et al.2000).

The MAM-E17 model also displays several behavioralcharacteristics reminiscent of cognitive impairments ob-served in schizophrenia, including deficits in classic para-digms such as prepulse inhibition and latent inhibition(Flagstad et al. 2005; Moore et al. 2006). MAM-E17 ratsshow impaired prefrontal cortical-dependent reversal andextra-dimensional shift learning on a bowl-digging task(Featherstone et al. 2007; Gastambide et al. 2012) and def-icits in hippocampal-dependent spatial memory on mazenavigation tasks (Gourevitch et al. 2004; Featherstoneet al. 2009; Gastambide et al. 2015). Impairments in

inhibitory control have also been observed as decreased ef-ficiency on a differential reinforcement of low-rate (DRL-20) task and nonsignificant trends toward increased prema-ture responding in the five-choice serial reaction time task(5-CSRTT) (Featherstone et al. 2007). Some patients diag-nosed with schizophrenia show potentially analogous cogni-tive impairments when challenged with computerized neuro-psychological tests (Hutton et al. 1998; Birkett et al. 2007;Leeson et al. 2009; Nuechterlein et al. 2015).

One cognitive paradigm that is particularly sensitivefor detecting reliable impairments in schizophrenia pa-tients is the continuous performance test (CPT)(Wohlberg and Kornetsky 1973; Buchanan et al. 1997).Typical versions of the CPT require subjects to visuallyattend to a single location on a monitor and report brieflypresented target stimuli amongst a stream of nontargetstimuli. Numerous cognitive/executive processes are re-quired for successful CPT performance including atten-tion (alerting, selective, or sustained), memory (targetsversus nontargets), inhibitory control (withholding inap-propriate responses), and processing speed (rapidly iden-tifying targets). Schizophrenia patients display impair-ments in the CPT paradigm that appear independent ofsymptom severity (Nuechterlein et al. 2015) as well asclinical (Wohlberg and Kornetsky 1973; Cornblatt et al.1997) and medication (Cornblatt et al. 1997; Keefe et al.2006; Nuechterlein et al. 2015) status. Despite consistentCPT impairments observed in schizophrenia and robustfrontostriatal alterations in MAM-E17 animals, behavioraleffects in the MAM-E17 model using rodent sustainedattention paradigms have been equivocal. Limited evi-dence of disrupted performance in MAM-E17 rats hasbeen detected in either the 5-CSRTT (Featherstone et al.2007) or the sustained attention task (without or includingdistractors (SAT/dSAT)) (Howe et al. 2015), even follow-ing extensive parameter manipulations taxing attentionaland impulsive processes.

The current study implemented a novel touchscreen-based rodent continuous performance task (rCPT) (Kimet al. 2015). The rCPT closely emulates the task structureand main cognitive requirements of common variants ofthe human CPT by incorporating a continuous successionof several luminance-matched target and nontarget imagesthat require both detection and discrimination as well asresponse inhibition on nontarget trials. We hypothesizedthat the rCPT might thus be more sensitive for detectingimpairments in attentional and/or executive function inthe MAM-E17 model. We further evaluated the impactof eight acute systemic treatments of compounds that af-fect a variety of neurotransmitter systems and have dem-onstrated potential to enhance cognitive performance inrodents and/or have efficacy for remediating impairmentsin clinical patient groups.

2838 Psychopharmacology (2017) 234:2837–2857

Methods

Animals

Male offspring of Sprague Dawley rat dams treated on E17with 24 mg/kg (doses calculated as salt) intraperitonealmethylazoxymethanol acetate (MAM-E17) or saline (sham)were generated at Charles River UK, Ltd. (Kent, UK). TheMAM-E17 treatment procedure was similar to that describedby Moore et al. (2006). Offspring were received 12–16 weeksafter birth. Rats were housed in groups of three to four percage under a reverse (12 h:12 h) light cycle (lights off at0700 hours) in a temperature- and humidity-controlled vivar-ium. After a ≥1-week acclimatization period, the animals’access to standard diet (standard rat chow; Nestlé Purina,St. Louis, MO, USA) was restricted. Animals were main-tained at ≈85–90% of their free-feeding weight throughouttesting. Water was available ad libitum except during operanttesting. The experiments used two cohorts of animals testedsequentially 1 year apart [cohort 1: MAM (n = 12) andSHAM (n = 7); cohort 2: MAM (n = 15) and SHAM(n = 16)]. All experiments were carried out in compliancewith the UK Animals (Scientific Procedures) Act 1986 andUK Animals (Scientific Procedures) Act 1986 AmendmentRegulations 2012.

Apparatus

The experiments used 18 Med Associates, Inc. modular oper-ant behavioral chambers (30.5 × 24.1 × 8.25 cm) modified toinclude a touch-sensitive monitor (Elo Touch Solutions, Inc.,Milpitas, CA, USA; see Mar et al. 2013; Oomen et al. 2013)controlled through in-house software (Visual Basic 2010Express .NET, Microsoft 2010; developed by A.C.M.). Eachchamber was located within a sound-attenuating box with afan installed for ventilation and to mask external noise. Therear wall of each chamber was equipped with a food magazine(lower center) connected to a pellet dispenser delivering45 mg rewards, a tone generator (upper corner), and a houselight (upper center). The food magazine was fitted with a traylight and a photocell beam detector for recording head entriesand reward collection.

Drugs

Drug dosing protocols were based on both in-house pilot ex-periments and previously published reports (Waters et al.2005; Morgan et al. 2007; Redrobe et al. 2012; Prickaertset al. 2012; Gastambide et al. 2012).

Sulpiride (Sigma-Aldrich, St. Louis, MO, USA), a dopa-mine D2/D3R antagonist, was mixed in sterile saline and ad-ministered intraperitoneally (i.p.) at 0, 1, 3, and 10 mg/kgdoses. Sulpiride has been demonstrated to reduce certain

positive and negative symptoms of schizophrenia but appearsto have limited efficacy against the cognitive impairments ofthe disorder (Soares et al. 2000).

Atomoxetine hydrochloride (Sigma-Aldrich, St. Louis,MO, USA), a noradrenaline reuptake inhibitor (Wong et al.1982), was mixed in sterile saline and injected i.p. at 0, 0.1,0.3, and 1.0 mg/kg doses. Atomoxetine can improve certainattentional measures in individuals with ADHD (Chamberlainet al. 2007; Maziade et al. 2009; Barry et al. 2009) as well asresponse inhibition in healthy volunteers (Chamberlain et al.2006).

LSN2463359 (Eli Lilly & Co., Ltd., Windlesham, UK), amGlu5R positive allo steric modulator (PAM), was suspendedin 1% carboxymethyl cellulose, 0.25% Tween 80, and 0.05%antifoam and administered per os (p.o.) via syringe at 0, 1.0,2.5, and 5.0 mg/kg doses. LSN2463359 has been demon-strated to restore cognitive inflexibility deficits in theMAM-E17 model (Gastambide et al. 2012).

RO4938581 (F. Hoffmann La Roche, Ltd., Basel,Switzerland), an inverse agonist at the GABA(A)α5R, wassuspended 0.3% Tween 80 and 0.05% antifoam and adminis-tered p.o. via gavage at 0, 0.1, 1.0, and 10.0 mg/kg doses.RO4938581 improves learning and memory in several rodentmodels (Ballard et al. 2009; Redrobe et al. 2012).

Modafinil (microionized; Eli Lilly &Co., Ltd., Windlesham,UK), an atypical stimulant and vigilance promoter, wassuspended in 10% (w/v) sucrose and administered p.o. via sy-ringe at 0, 8, 32, and 64 mg/kg doses. Modafinil has cognitive-enhancing and vigilance-promoting properties in various psy-chiatric disorders (Randall et al. 2004, 2005; Turner et al. 2004;Hunter et al. 2006; Dean et al. 2011; Schmaal et al. 2013).

ABT-594 (AbbVie, Inc., IL, USA), a potent agonist atthe α4β2nAChR (Donnelly-Roberts et al. 1998), wasmixed in sterile saline and injected i.p. at 0, 1.9, 5.9,and 19 μg/kg doses. ABT-594 has been shown to improveaspects of performance on preclinical, sustained attentiontasks (McGaughy et al. 1999; Mohler et al. 2010; Howeet al. 2010).

Donepezil hydrochloride (Sigma-Aldrich, St. Louis, MO,USA), an acetylcholinesterase inhibitor, was mixed in sterilesaline and injected i.p. at 0, 0.1, 0.3, and 1.0 mg/kg doses.Donepezil can improve attentional measures in individualswith Alzheimer’s disease (Sahakian and Coull 1993; Foldiet al. 2005).

EVP-6124 (EnVivo Pharmaceuticals, Inc., Watertown,MA, USA), a partial α7nAChR agonist, was suspended in10% sucrose in deionized water vehicle and administeredp.o. via syringe at 0, 0.3, 1.0, and 3.0 mg/kg doses. EVP-6124 showed some indications for therapeutic potential(Olincy et al. 2006; Freedman et al. 2008; Liebermanet al. 2013; Preskorn et al. 2014) but failed to showcognitive-enhancing effects in larger clinical trials(Fidler 2016).

Psychopharmacology (2017) 234:2837–2857 2839

Behavioral procedure

Initial training on the rCPT was comprised of four stages. Inthe first stage, rats were trained to attend, approach, and toucha solid white square stimulus (7 × 7 cm) presented centrally onthe touchscreen. Each white square stimulus was presented fora maximum of 10-s stimulus duration (SD). A 2-s inter-stim-ulus interval (ISI) was given between stimulus presentationsin which only a white frame outlining the location of theresponse window was visible. Screen touches made withinthe response window either while the stimulus was being pre-sented or in less than 500ms following stimulus removal (SD +500 ms = limited hold (LH) period) were designated as hits(correct responses). Following hits, the stimulus (if present)was removed immediately from the screen, the magazine lightwas illuminated, and a single 45 mg food reward pellet wasdelivered to the magazine. Reward collection extinguished themagazine light and initiated the next trial ISI. Screen toucheswithin the response window during the ISI reset the interval,thus delaying the onset of the next stimulus presentation.Stimuli that were not touched within the LH period were clas-sified as misses (stimulus omissions). Sessions were 45–60 minin duration. Criterion for phase 1 was defined as earning 100rewards within a session across two consecutive sessions.

In stage 2, the white square was replacedwith a novel target(S+) stimulus and the SDwas reduced to 2 s (LH = 2.5 s). Thenew S+ was either a horizontal or vertical line stimulus,counterbalanced across both MAM and SHAM groups.Additionally, a brief ingestion delay (ID) period of 5 s wasintroduced following reward collection to permit the animaltime to consume the food pellet before re-engaging in the task.The ISI prior to the next stimulus began immediately aftertermination of the ID. Animals were trained for one to twosessions on stage 2.

In stage 3, a novel nontarget (S-) stimulus was introduced tothe stimulus set. The new S− was either a vertical or horizontalline stimulus—whichever was different from the S+ that wascounterbalanced and randomly assigned in stage 2. The SD ofboth S+ and S− stimuli was equated at 2 s (LH= 2.5 s), and therewas a 50% probability of either stimulus appearing on any givennormal trial. The ISI was increased to 5 s in stage 3. Touchingthe S− stimulus during the LH period was designated as a falsealarm (incorrect response) and resulted in immediate removal ofthe stimulus (if present) and initiation of a correction-trial ISI.On all correction trials, the S−was the only stimulus presented.These correction trials were introduced for a similar reason asthe inclusion of ISI touch resetting (see stage 1—to discouragenonselective responding to stimuli and the central screen whenthe target stimulus is absent). Withholding responses to the S−during the LH period was designated as a correct rejection andinitiated a normal-trial ISI. All other parameters were identicalto those of stage 2. Animals were trained for seven to eightsessions on stage 3.

A flowchart overview of the rCPT trial structure for thefourth and final training stage is depicted in Fig. 1a. In thisstage, three additional S– stimuli were introduced to the stim-ulus set. All stimuli were luminance matched in an attempt toequate their salience for simple brightness detection. The SDof S+ and S– stimuli was set to be variable (0.5–1.5 s) but witha fixed LH of 2 s. The LH remained at 2 s after stimulus onsetto maintain the average event rate and to ensure that animalshad equal opportunity to respond to each stimulus. As withstage 3, the probability of presentation of either an S+ or an S–stimulus on normal trials was equal (50%). The ISI was alsoset to be variable (3–7 s). The variable SD and ISI were intro-duced to increase the attentional load of the task. Other pa-rameters were the same as stage 3, with touches to any of theS– stimuli resulting in a correction trial in which an S– waspresented randomly from the set of four. All animals weretrained a minimum of 20 sessions on stage 4 to establishreliable rCPT performance prior to drug testing. Animals weretrained on the rCPT for daily sessions, 5–7 days a week.

Experimental design: drug testing

The effects of different compounds were investigated usingstage 4 performance parameters. Drugs were administered fol-lowing a diagram-balanced Latin square design (Cardinal andAitken 2006). Each drug Latin square experiment consisted offour test sessions with one dose per session. Each dosing ses-sion was separated by at least 48 h with no-drug baseline ses-sions run in between. A typical minimum interval of 1 week,with several no-drug baseline retention sessions being run, wasgiven between different drug treatments. The order of the drugsadministered to the two cohorts is illustrated in Fig. 1b. Incohort 1, the effects of EVP-6124, LSN2463359, modafinil,and RO4938581 were examined on rCPT performance. Thesession duration was 60 min in cohort 1 but was restrictedto 45 min in cohort 2 as it had been determined that theadditional 15 min did not greatly alter the pattern of rCPTresults. In cohort 2, RO4938581, donepezil, ABT-594,atomoxetine, and sulpiride were tested. RO4938581 wasassessed in both cohorts to control for cross-batch reproduc-ibility of pharmacological effects. Prior to the Latin squareexperiment for ABT-594, a preliminary dose (19 μg/kg)was administered in a crossover design during rCPT testingto permit acclimation to the initial effects of the drug(Mohler et al. 2010). All compounds were administered30 min prior to rCPT testing.

Data analysis

Only the first 45 min of each session was analyzed tobetter compare data across cohorts. However, no signif-icant differences in performance variables were observed


regardless of whether a 45- or 60- min session wasanalyzed in cohort 1.

Behavioral parameters Analogous to human CPTs, a re-sponse to the target (S+) was classified as a hit, failure torespond to the target was classified as a miss, withholdingfrom responding to a nontarget (S–) was recorded as acorrect rejection (CR), and responding to a nontargetwas classified as a false alarm. Responses to the screenduring the ISI (when no stimulus was present) were codedas ISI touches. For each animal, hit rate (HR) was calcu-lated as the number of hits as a proportion of the totalnumber of CS+ presentations (hits / (hits + misses).False alarm rate (FAR) was calculated as the number offalse alarms as a proportion of the total number of CS–presentations (FA / (FA + CR)). The discrimination

sensitivity index was calculated as (Macmillan andCreelman 2004)

d0 ¼ z hit rateð Þ−z false alarm rateð Þ

with higher values indicating better discrimination sensitiv-ity across the session. The response criterion was calculated as

c ¼ −0:5 z hit rateð Þ þ z false alarm rateð Þð Þ

with larger values indicating decreased overall responding toboth the target and nontarget stimuli. An identical pattern ofsignificant effects was obtained when nonparametric indicesof sensitivity and response bias were calculated (Stanislaw andTodorov 1999), which do not depend on assumptions such asnormal distributions and equal variances for the discriminability

a

b

Fig. 1 Trial structure of the rCPT and timeline of drug administration forthe two cohorts of MAM-E17 and sham control rats. a Schematic dia-gram of the trial structure of the rCPT. Each session consists of a series oftrials. Each trial begins (see Trial Starts box) with a variable inter-stimulus interval (ISI) = 3–7 s in which no stimuli are present. A responseto the screen during the ISI is coded as an ISI touch and reinitiates the ISI.Following the complete ISI, either a designated target stimulus (S+) orone of four nontarget stimuli (S–) is presented individually within a cen-tral response window on a touch-sensitive monitor. Each stimulus is pro-grammed to have a variable stimulus duration (SD) = 0.5–1.5 s. Onnormal trials, if an S+ is displayed, a touch to the response windowduring a limited hold period following stimulus onset (LH = 2 s) is clas-sified as a hit and leads to removal of the visual stimulus and reward

delivery. Following reward collection, the normal-trial ISI prior to thenext stimulus is initiated after a brief ingestion delay (ID) = 5 s that affordsthe animal time to consume the reward. Failure to respond to the S+stimulus within the LH is recorded as a miss and initiates the next nor-mal-trial ISI. If an S– is displayed, omitting a response within the LH isclassified as a correct rejection and initiates the next normal-trial ISI. Aresponse-window touch during the LH of S– presentation is classified as afalse alarm and causes removal of the visual stimulus and onset of acorrection-trial ISI. Correction trials—in which a randomly selected S–is presented—are repeated until a successful correct rejection is complet-ed. b Timeline illustrating the order of the drug administration studies incohorts 1 and 2 of MAM-E17 and sham control rats


of targets and nontargets. The primary dependent variables weresensitivity index (d′), response criterion (c), hit rate, false alarmrate, and ISI touches. Correct response latency, incorrect re-sponse latency, and reward retrieval latency were also assessed.

Independent sample t tests were used to compareMAM andsham performance in stage 1 (sessions to criterion), stage 2(average hit rate), and stage 3 (d′). To assess the robustnessof stage 4 rCPT performance in MAM and sham control ani-mals, the average performance of two to three drug-free base-line sessions for each measure was analyzed at time pointsimmediately prior to each Latin square experiment. Only fourof five of the equivalent time points were analyzed in cohort 2for comparison with cohort 1. A three-way mixed ANOVAwas used in which cohort [1, 2] and group [MAM, sham] wereincluded as between-subjects factors and time point [1, 2, 3, 4]was included as a within-subjects factor. To further examinethe stability of the MAM versus sham differences, each cohortwas also analyzed separately at each time point from cohorts 1and 2 with group as a factor. Effect sizes for the difference in d′between MAM and sham rats at each time point were alsocalculated. The vehicle treatment trials from each Latin squareexperiment were also analyzed in a similar manner.

Data from each drug experiment were analyzed using re-peated measures ANOVA, with group (between-subjects, two

levels of dose: sham versus MAM treatment) and drug dose(within-subjects, four levels of dose) as independent variables.Significant main effects of drug were followed by post hoccomparisons against vehicle using the Šidák adjustment formultiple comparisons. Significant interactions were investigat-ed further by analysis of simple main effects and Šidák-corrected pairwise comparisons. Data were analyzed usingSPSS Statistics 21 (IBM Corp., Armonk, NY, USA). We addi-tionally confirmed all of our repeatedmeasures ANOVA resultsusing an alternate statistical model (replicated Latin square de-sign with a between-subjects factor) that accounted for day andorder effects (Cardinal and Aitken 2006). Moreover, we foundno significant differences in our results when correction trialswere either included or excluded from the analysis.

Results

Effect of MAM-E17 on rCPT performance

Across the two cohorts of animals and eight drug studiesspanning several months, the MAM-E17 rats showed severalhighly consistent and selective differences in rCPT perfor-mance relative to sham controls (Fig. 2d, Table 1). During

Fig. 2 rCPT acquisition data and long-term stability of baseline taskperformance in MAM-E17 and sham control animals. a Stage 1 no dif-ferences between MAM-E17 and sham groups on the number of sessionsto criterion. b Stage 2 no differences between MAM-E17 and shamgroups on mean hit rate. c Stage 3 no differences between MAM-E17and sham animals on performance sensitivity, d′, averaged across the last3 days of testing. d Stage 4 mean d′ in the rCPT at four baseline time

points for cohorts 1 and 2. Each time point comprises two to three ses-sions prior to commencing a Latin square drug study. MAM-E17 ratsshow stable deficits in d′ across the four time points spanning severalmonths. There were no significant effects of cohort or cohort × groupinteractions. Data are expressed as mean ± SEM. Hash tags denote sig-nificant main effects of group (#p < 0.05; ##p < 0.01)


rCPT acquisition, there were no significant differences be-tween MAM and sham control rats or between cohorts 1 and2 in terms of the number of sessions to criterion in stage 1(Fig. 2a), the hit rate or number of pellets earned per sessionduring stage 2 (Fig. 2b), or the sensitivity (d′) during stage 3(Fig. 2c). However, when examining stage 4 rCPT perfor-mance across baseline sessions prior to each Latin square drugstudy, MAM animals exhibited significantly decreased d′(Fig. 2d; F[1,46] = 15.815, p < 0.001) and significantly in-creased false alarm rate (F[1,46] = 18.160, p < 0.001) andnumber of ISI screen touches (F[1,46] = 7.030, p = 0.011)relative to sham controls in both cohorts 1 and 2. There wereno significant main effects of cohort or group × cohort inter-actions for any of the measures, suggesting that the MAM-E17 model induced specific and robust performance differ-ences in the rCPT relative to sham controls. When each cohortand time point was analyzed separately, the effects on both d′and false alarm rate were significant at three of four time pointsin cohort 1 (all p values <0.05) and at every time point in cohort2 (all p values <0.05), highlighting the stability of the observeddifferences between MAM-E17 and sham rats across manymonths. The effect sizes of the differences in d′ betweenMAM-E17 and shams at each time point were large, rangingfrom 0.68 to 1.34 in cohort 1 (overall = 1.23) and from 1.10 to1.37 in cohort 2 (overall = 1.27).

A similar pattern was observed when measures were aver-aged across only the vehicle treatment sessions in both cohorts(four and five sessions in cohorts 1 and 2, respectively). MAM-E17 rats showed significantly lower performance sensitivity, d′,relative to sham control animals (F[1,46] = 13.480, p = 0.001).In addition, MAM-E17 rats displayed significantly higher FAR(F[1,46] = 13.189, p = 0.001) and ISI touches (F[1,46] = 7.746,p = 0.008). There were no MAM-E17 effects on any othermeasures (HR, c, as well as correct, incorrect, and reward col-lection latencies). There were no significant main effects of co-hort or group × cohort interactions for any of themeasures, again

suggesting that the observed effects of the MAM-E17 model onrCPT performance were specific, stable, and reproducible.

Effect of acute drug treatments on rCPT performance

All of the drug effects on the primary rCPT performance in-dices (d′, HR, FAR, c, and ISI touches) are presented inFigs. 3, 4, and 5 and summarized in Table 2 andSupplementary Table S1. Latency measures (correct response,incorrect response, and reward collection latency) are present-ed in Table 3. Only significant and near-significant effects arehighlighted below. Results on nonsignificant measures areincluded in the Supplement.

Sulpiride Sulpiride (0–30 mg/kg) dose-dependently and se-lectively reduced the elevated FAR of MAM animals on therCPT (Fig. 3b, Table 2). It also dose-dependently lowered over-all levels of responding (increased c with decreased HR, FAR,and ISI touches) and increased correct response latency (Fig.3a, b, Table 3). One sham animal was excluded from analysisdue to a data acquisition error during the vehicle dose session.

For FAR, there was both a significant main effect of dose(dose: F[3,84] = 10.203, p < 0.001; H-F corrected) and a sig-nificant dose × group interaction (dose × group:F[3,84] = 4.119, p = 0.016; H-F corrected). Post hoc testsconfirmed that sulpiride significantly reduced FAR selectivelyin MAM-E17 animals at the highest 30 mg/kg dose relative tovehicle (p < 0.05), such that MAM-E17 animals no longerdisplayed significantly elevated FAR relative to sham controlsat the 30 mg/kg dose (p = 0.206 where p < 0.05 at all othersulpiride doses). There were no significant FAR changes insham control animals under sulpiride treatment. There werealso significant main effects of sulpiride dose on HR (dose:F[3,84] = 7.614, p < 0.001; dose × group: F[3,84] = 0.149,p = 0.900; H-F corrected), c (dose:F[3,84] = 10.921, p < 0.001;dose × group: F[3,84] = 0.994, p = 0.385; H-F corrected), andISI touches (dose: F[3,84] = 3.977, p < 0.014; dose × group:F[3,84] = 0.867, p = 0.450; H-F corrected). Within-subjectscontrasts revealed significant linear relationships between as-cending sulpiride doses and reductions in HR (F[1,28] = 9.623,p = 0.004) and ISI touches (F[1,28] = 7.947, p = 0.009), as wellas with increases in c (F[1,28] = 15.014, p < 0.001). Post hoctests showed that the highest 30 mg/kg sulpiride dose signifi-cantly reduced HR and ISI touches and significantly increasedresponse criterion, c, relative to vehicle-treated animals acrossboth MAM and sham control groups (all p < 0.05).

There was a significant main effect of sulpiride on correctresponse latency (dose: F[3,84] = 8.145, p < 0.001; dose ×group: F[3,84] = 1.610, p = 0.193), with linear dose-dependent increases (F[1,29] = 12.451, p = 0.001) such thatthe highest 30 mg/kg sulpiride dose resulted in longer correctresponse latencies relative to the vehicle treatment (p < 0.05).

Table 1 Summary of the MAM-E17 rat model phenotype relative tothe sham controls on rCPT performance variables

Performance measure MAM-E17 phenotype

Hit rate –

False alarm rate ↑

d′ ↓

c –

ISI touches ↑

Correct latency –

Incorrect latency –

Retrieval latency –

↑ = increased in the MAM-E17 model; ↓ = decreased in the MAM-E17model; – = not affected in the MAM-E17 model


Atomoxetine Atomoxetine (0–1 mg/kg) had significant sup-pressant effects on all of the primary response measures (d′, c,HR, FAR, ISI touches) and increased correct and incorrectresponse latencies in the rCPT (Fig. 3c, d, Tables 2 and 3).

Atomoxetine caused dose-dependent decreases in perfor-mance sensitivity, d′ (dose: F[3,87] = 2.896, p = 0.040; dose× group: F[3,87] = 0.408, p = 0.748), HR (dose:F[3,87] = 11.517, p < 0.001; dose × group: F[3,87] = 0.533,

p = 0.661), FAR (dose: F[3,87] = 6.796, p < 0.001; dose ×group: F[3,87] = 0.483, p = 0.695), and ISI touches (dose:F[3,87] = 8.729, p < 0.001; dose × group: F[3,87] = 1.830,p = 0.148). It also significantly increased response criterion, c(dose: F[3,87] = 12.473, p < 0.001; dose × group:F[3,87] = 0.462, p = 0.709).Within-subjects contrasts revealedsignificant linear effects on all of these measures, showing thatincreasing atomoxetine doses were associated with decreases

Fig. 3 Effect of sulpiride, atomoxetine, LSN2463359, and RO4938581on rCPT performance inMAM-E17 and sham control rats. All significantgroup differences are relative to sham controls, and significant drug ef-fects are in comparison to the vehicle treatment. For all compounds,MAM-E17 rats exhibited a dose-independent decrease in d′ relative toshams (a, c, e, g) a, b Sulpiride. a Sulpiride (30 mg/kg) increased re-sponse criterion (c) in MAM-E17 animals without affecting d′. bSulpiride (30 mg/kg) reduced the elevated FAR selectively in MAM-E17 animals and caused reductions in HR and ISI touches across bothMAM-E17 and sham animals. c, d Atomoxetine. c Atomoxetine in-creased c at both 0.3 and 1.0 mg/kg doses and reduced d′ at a 1.0 mg/kg dose. d Atomoxetine decreased HR and d′ at 0.3 and 1.0 mg/kg andreduced FAR at 1.0 mg/kg. MAM-E17 animals showed a dose-independent increase in FAR relative to shams. e, f LSN2463359. eLSN2463359 (5 mg/kg) increased c across both groups of animals.

MAM-E17 animals showed a dose-independent decrease in d′ comparedto shams. f LSN2463359 (5 mg/kg) reduced HR and ISI touches in bothMAM-E17 and sham groups. MAM-E17 animals showed a dose-independent increase in FAR. g, h RO4938581. g RO4938581 (10 mg/kg) increased c in both groups of rats. MAM-E17 animals showed a dose-independent decrease in d′ relative to sham controls. h RO4938581(10 mg/kg) decreased HR, FAR, and ISI touches in both groups.MAM-E17 animals showed a dose-independent increase in FAR com-pared to sham controls. Gray shading with asterisks denotes the maineffect of drug dosewith a significant difference from vehicle.Red shadingwith asterisks denotes group × dose interaction with a significant differ-ence from vehicle (*p < 0.05; **p < 0.01; ***p < 0.0001). Hash tagsdenote significant main effects of group (#p < 0.05; ##p < 0.01;###p < 0.0001)


in d′ (F[1,29] = 9.033, p = 0.005), HR (F[1,29] = 23.855,p < 0.001), FAR (F[1,29] = 12.977, p = 0.001), and ISI touches(F[1,29] = 29.132, p < 0.001), as well as with increases in c(F[1,29] = 22.471, p < 0.001). Post hoc tests showed that,relative to the vehicle treatment, the highest 1 mg/kg dosesignificantly decreased d′, HR, and FAR and that both 0.3and 1 mg/kg doses led to significantly fewer ISI touches andan increased response criterion, c (all p values <0.05).

There were significant main effects of atomoxetine dose onboth correct response latency (dose:F[3,87] = 12.295, p < 0.001;dose × group:F[3,87] = 0.517, p = 0.672) and incorrect responselatency (dose: F[3,87] = 8.692, p < 0.001; dose × group:F[3,87] = 2.512, p = 0.075; H-F corrected).Within-subjects con-trasts revealed significant linear effects indicating a dose-dependent slowing of correct (F[1,29] = 22.472, p < 0.001)and incorrect (F[1,29] = 13.772, p < 0.001) response latencies.Post hoc comparisons showed that the highest 1 mg/kg dose ofatomoxetine resulted in prolonged response latencies relative tothe vehicle treatment (all p values <0.05).

LSN2463359 LSN2463359 (0–5 mg/kg) dose-dependentlyreduced levels of rCPT responding, as evidenced by sig-nificant changes in HR, c, and ISI touch parameters acrossboth MAM-E17 and sham control groups (Fig. 3e, f,Tables 2 and 3). Two sham animals did not receive oneof the appropriate LSN2463359 dose levels and were thusexcluded from statistical analysis.

There were significant main effects of LSN2463359 doseon HR (dose: F[3,45] = 4.893, p = 0.017; dose × group:F[3,45] = 0.023, p = 0.970; H-F corrected), c (dose:F[3,45] = 5.729, p = 0.016; dose × group: F[3,45] = 0.357,p = 0.638; H-F corrected), and ISI touches (dose:F[3,45] = 3.681, p = 0.019; dose × group: F[3,45] = 0.197,p = 0.898). Within-subjects contrasts revealed significant pat-terns such that increasing LSN2463359 doses were linearlyassociated with decreases in HR (F[1,15] = 8.291, p = 0.011)and ISI touches (F[1,15] = 6.921, p = 0.019), as well as withlinear increases in c (F[1,15] = 7.919, p = 0.013). Post hoccomparisons showed that the highest 5 mg/kg LSN2463359dose significantly reduced HR and ISI touches and increasedresponse criterion, c, relative to vehicle-treated animals acrossboth MAM-E17 and sham control groups (all p values <0.05).There was also a nonsignificant trend for higher LSN2463359doses to reduce FAR (dose:F[3,45] = 2.395, p = 0.081; dose ×group: F[3,45] = 0.159, p = 0.923).

There was a significant dose × group interaction for rewardcollection latency (dose: F[3,42] = 2.110, p = 0.144; dose ×group: F[3,42] = 3.609, p = 0.045; H-F corrected), but posthoc tests revealed no significant pairwise comparisons.

RO4938581 RO4938581 (0–10 mg/kg) dose-dependently re-duced overall levels of responding on the rCPT, as indicated byan increased c parameter; decreased HR, FAR, and ISI touches;and increased correct response latencies (Fig. 3g, h, Tables 2and 3). The effects were highly consistent across both cohorts

Fig. 4 Effect of modafinil and ABT-594 on rCPT performance in shamand MAM-E17-treated rats. All significant group differences are relativeto sham controls, and significant drug effects are in comparison to thevehicle treatment. a, b. Modafinil. a At 8 mg/kg, modafinil increased d′selectively in the sham group without affecting d′ in MAM-E17 animals.At 64 mg/kg, modafinil decreased d′ and response criterion (c) acrossboth groups. b At 64 mg/kg, modafinil increased FAR and ISI touchesand in both groups. c, dABT-594. c ABT-594 caused group-independentdecreases in c at both 5.9 and 19.4 μg/kg doses. MAM-E17 animals

showed dose-independent reductions in d′ relative to controls. d ABT-594 (19.4μg/kg) also increased FAR, HR, and ISI touches in both groups.MAM animals showed dose-independent increases in FAR and ISItouches.Gray shading with asterisks denotes the main effect of drug dosewith a significant difference from vehicle. Red shading with asterisksdenotes group × dose interaction with a significant difference from vehi-cle (*p < 0.05; **p < 0.01; ***p < 0.0001). Hash tags denote significantmain effects of group (#p < 0.05; ###p < 0.0001)


of animals. From both cohorts 1 and 2, four of 27 MAM-E17rats and one of 23 sham control animals had data missing fromone of the dosing sessions and were thus excluded from statis-tical analysis.

There were significant main effects of RO4938581 dose onHR (dose: F[3,129] = 7.553, p < 0.001; dose × group:F[3,129] = 0.011, p = 0.998), FAR (dose: F[3,129] = 10.134,p < 0.001; dose × group: F[3,129] = 1.570, p = 0.200), c (dose:F[3,126] = 9.395, p < 0.001; dose × group: F[3,126] = 0.053,p = 0.984), and ISI touches (dose: F[3,129] = 7.172, p < 0.001;dose × group: F[3,129] = 0.845, p = 0.457; H-F corrected).Within-subjects contrasts revealed significant linear patterns,with increasing RO4938581 doses associated with reductionsin HR (F[1,43] = 11.533, p = 0.001), FAR (F[1,43] = 28.607,p < 0.001), and ISI touches (F[1,43] = 23.687, p < 0.001), aswell as with increases in c (F[1,42] = 19.849, p < 0.001). Post

hoc tests showed that, compared to the vehicle treatment, thehighest 10 mg/kg RO4938581 dose significantly reduced HR,FAR, and the number of ISI touches and increased responsecriterion, c, across both MAM-E17 and sham control groups(all p values <0.001).

There was a significant main effect of RO4938581 dose oncorrect response latency (dose:F[3,126] = 4.580, p= 0.004; dose× group: F[3,126] = 0.358, p = 0.783), with a significant within-subjects contrast indicating that ascending doses were linearlyrelated with increases in correct response latencies(F[1,42] = 15.055, p < 0.001). Post hoc comparisons showedthat the 10 mg/kg RO4938581 dose significantly increased cor-rect latencies relative to the vehicle treatment (p values <0.05).

To examine the consistency of RO4938581 effects acrosscohorts, the data were analyzed adding cohort as a fixed factor.Across all dependent measures, there was only a single

Table 2 Summary ofpharmacological effects on themain rCPT performance variables

Drug Hit rate False alarm rate d′ c ISI touches

Sulpiride ↓ ↓a – ↑ ↓

Atomoxetine ↓ ↓ ↓ ↑ ↓

LSN2463359 ↓ \ – ↑ ↓

RO4938581 ↓ ↓ – ↑ ↓

Modafinil – ↑ ↑ (low dose)b; ↓ (high dose) ↓ ↑

ABT-594 ↑ ↑ – ↓ ↑

Donepezil – – / – –

EVP-6124 – – – – –

↓ = significant decrease; ↑ = significant increase; \ = trend toward decrease; / = trend toward increase; – = no effecta Effect in the MAM-E17 model onlyb Effect in the sham controls only

Fig. 5 Effect of donepezil and EVP-6124 on rCPT performance in shamand MAM-E17-treated rats. a, b Donepezil. a There was a near-significant improvement in d′ by donepezil (0.1 mg/kg) across bothgroups of animals. MAM-E17 animals exhibited dose-independent de-creases in d′. b No effects of EVP-6124 on HR, FAR, or ISI touches.

MAM-E17 animals showed a dose-independent increase in FAR. c, d.EVP-6124. c, d No effects of EVP-6124 on any principal rCPT perfor-mance indices.Gray-shaded asterisk denotes the main effect of drug dosewith a significant difference from vehicle. Hash tags denote a significantmain effect of group (#p < 0.05; ##p < 0.01)


significant effect of cohort: a group × dose × cohort interactionon the response criterion index, c (F[3,120] = 3.791,p = 0.012), in which sham control animals from cohort 1had a decreased c parameter when they were administered1 mg/kg RO4938581 as compared to similarly treated animalsfrom cohort 2 (p < 0.05). Apart from this isolated effect, there

were no significant main effects of cohort or interactions in-volving cohort × group, cohort × dose, or cohort × group ×dose for any of the rCPT performance measures. The generalabsence of any statistical effects of cohort on themain findingsof RO4938581 underscores the reliability and reproducibilityof these drug effects on the rCPT paradigm.

Table 3 Response and retrieval latencies of MAM-E17 and sham animals in the rCPT during acute systemic pharmacological challenges

Correct response latency Incorrect response latency Reward retrieval latency

Drug Sham MAM Sham MAM Sham MAM

Modafinil

Vehicle 808 ± 46 827 ± 32 727 ± 48 768 ± 36 1451 ± 118 1424 ± 114

8 790 ± 47 813 ± 25 774 ± 55 746 ± 35 1960 ± 410 1403 ± 128

32 776 ± 59 787 ± 23 727 ± 42 683 ± 15 1460 ± 134 1781 ± 365

64 750 ± 56 788 ± 28 706 ± 37 739 ± 29 1411 ± 147 1442 ± 184

ABT-594

Vehicle 852 ± 31 784 ± 33 649 ± 29 672 ± 30 1531 ± 110 1953 ± 608

1.9 885 ± 35 773 ± 46 661 ± 33 669 ± 36 2096 ± 436 1445 ± 125

5.9 839 ± 35 769 ± 36 694 ± 39 641 ± 28 1555 ± 91 1503 ± 253

19.4 831 ± 35 712 ± 35 691 ± 40 611 ± 32 1633 ± 140 1355 ± 166

LSN2463359

Vehicle 744 ± 71 798 ± 37 665 ± 88 741 ± 35 1454 ± 102 1561 ± 142

1.0 767 ± 84 811 ± 33 767 ± 104 704 ± 25 2316 ± 786 1382 ± 45

2.5 782 ± 88 864 ± 37 784 ± 96 742 ± 28 1468 ± 99 1464 ± 89

5.0 780 ± 95 855 ± 34 766 ± 97 786 ± 26 1625 ± 101 1498 ± 78

RO4938581

Vehicle 874 ± 29 838 ± 26 762 ± 31 722 ± 29 2005 ± 338 4830 ± 3450

0.1 886 ± 27 841 ± 34 715 ± 28 749 ± 32 5205 ± 3324 1887 ± 267

1.0 928 ± 29 850 ± 31 794 ± 28 720 ± 30 2020 ± 382 1610 ± 109

10.0 924 ± 29 897 ± 32 735 ± 26 722 ± 24 3442 ± 1233 4100 ± 1705

Donepezil

Vehicle 871 ± 38 798 ± 37 737 ± 31 670 ± 31 1410 ± 85 1633 ± 228

0.1 857 ± 33 785 ± 35 680 ± 30 683 ± 33 1959 ± 435 1495 ± 144

0.3 823 ± 31 800 ± 48 717 ± 43 711 ± 30 1592 ± 180 1805 ± 320

1.0 867 ± 26 793 ± 52 704 ± 44 689 ± 29 1897 ± 417 1584 ± 171

Atomoxetine

Vehicle 893 ± 31 815 ± 31 619 ± 39 700 ± 23 1835 ± 371 1654 ± 276

0.1 912 ± 32 814 ± 40 698 ± 35 663 ± 27 2311 ± 535 1621 ± 199

0.3 909 ± 31 861 ± 43 696 ± 40 668 ± 27 2200 ± 580 1490 ± 113

1.0 1006 ± 32 926 ± 42 757 ± 42 791 ± 40 1537 ± 64 1568 ± 103

Sulpiride

Vehicle 918 ± 35 796 ± 36 741 ± 54 691 ± 33 1835 ± 371 1654 ± 276

0.1 908 ± 37 846 ± 37 676 ± 45 684 ± 31 2311 ± 535 1621 ± 199

0.3 892 ± 33 844 ± 33 695 ± 51 737 ± 28 2200 ± 580 1490 ± 113

1.0 977 ± 36 900 ± 33 672 ± 46 705 ± 34 1537 ± 64 1568 ± 103

EVP-6124

Vehicle 868 ± 54 896 ± 44 655 ± 49 670 ± 50 1528 ± 144 1405 ± 104

0.3 901 ± 43 910 ± 37 666 ± 132 726 ± 40 3337 ± 1766 1421 ± 89

1.0 946 ± 65 927 ± 44 477 ± 73 696 ± 40 2187 ± 678 1355 ± 61

3.0 851 ± 70 917 ± 43 720 ± 43 702 ± 46 16,638 ± 15,164 1587 ± 302

For the significant effects of the drug dose denoted by color with statistical significance highlighted by color code, please see the legend

Sig. increase

from vehicle

0.01

0.001

0.05 Sig. decrease

from vehicle

p <


Modafinil Modafinil (8 mg/kg) selectively improved perfor-mance sensitivity, d′, in sham control rats (Fig. 4a, Table 2).Modafinil (0–64 mg/kg) also dose-dependently increasedlevels of responding within the rCPT, as demonstrated bydecreases in c, increases in FAR and ISI touches, and linearreductions in correct response latency with increasingmodafinil dose (Fig. 4a, b, Tables 2 and 3).

For performance sensitivity, d′, there was both a significantmain effect of dose (F[3,51] = 4.396, p = 0.008) and a signif-icant dose × group interaction (F[3,51] = 3.834, p = 0.015).Post hoc tests confirmed that low-dose (8 mg/kg) modafinilselectively enhanced d′ in sham control animals (increased d′relative to sham vehicle, p < 0.05). Post hoc comparisons with-in the main effect of dose also showed that, relative to vehicle,the highest 64 mg/kg modafinil dose reduced d′ performanceacross both MAM-E17 and sham control animals (p < 0.05).There were significant main effects of modafinil dose on FAR(dose: F[3,51] = 12.304, p < 0.001; dose × group:F[3,51] = 2.152, p = 0.105), c (dose: F[3,51] = 6.511,p = 0.001; dose × group: F[3,51] = 0.747, p = 0.529), and ISItouches (dose: F[3,51] = 8.480, p < 0.001; dose × group:F[3,51] = 0.184, p = 0.907). Within-subjects contrasts revealedsignificant linear relationships, with an increasing modafinildose associated with increases in FAR (F[1,17] = 35.048,p < 0.001) and ISI touches (F[1,17] = 17.681, p < 0.001), aswell as a decreases in response criterion, c (F[1,17] = 20.678,p < 0.001). Post hoc tests showed that, relative to the vehicletreatment, the highest 64 mg/kg modafinil dose significantlyelevated the FAR and number of ISI touches and decreased cacross all animals (all p values <0.05).

Within-subjects trend analysis also revealed a significantlinear relationship in which higher modafinil doses were asso-ciated with faster correct response latencies (F[1,17] = 8.367,p < 0.010).

ABT-594 ABT-594 (0–19.4 μg/kg) dose-dependently in-creased overall levels of responding by decreasing the c pa-rameter and increasing HR, FAR, and ISI touches. ABT-594also dose-dependently decreased correct response latencies(Fig. 4c, d, Tables 2 and 3).

In the acclimation to ABT-594 dosing using a crossoverdesign (data not shown), there was a significant group × doseinteraction (F[1,29] = 4.932, p < 0.05) observed for ISItouches, in which 19.4 μg/kg ABT-594 increased ISI touchesselectively in MAM rats.

In the Latin square, there were significant main effects ofABT-594 dose on HR (dose: F[3,87] = 5.897, p = 0.001; dose× group: F[3,87] = 0.422, p = 0.724), FAR (dose:F[3,87] = 17.807, p < 0.001; dose × group: F[3,87] = 0.238,p = 0.870), c (dose:F[3,87] = 15.072, p < 0.001; dose × group:F[3,87] = 0.276, p = 0.842), and ISI touches (dose:

F[3,87] = 13.432, p < 0.001; dose × group: F[3,87] = 2.552,p = 0.083; H-F corrected). Within-subjects contrasts revealedsignificant linear increases in HR (F[1,29] = 13.385,p = 0.001), FAR (F[1,29] = 42.314, p < 0.001), and ISItouches (F[1,29] = 26.450, p < 0.001), as well as a significantlinear decrease in c (F[1,29] = 32.801, p < 0.001) in responseto increasing ABT-594 doses. Post hoc comparisons showedthat the highest 19.4 μg/kg ABT-594 dose significantly in-creased HR, FAR, and ISI touches, and both the 5.9 and19.4 μg/kg doses decreased response criterion (c) relative tovehicle-treated animals (all p values <0.05).

ABT-594 dose-dependently speeded correct response la-tencies (dose: F[3,87] = 4.698, p = 0.004; dose × group:F[3,87] = 1.362, p = 0.260) where within-subjects contrastsrevealed a linear decrease in latencies across ascending doses(F[1,29] = 8.606, p = 0.006). Post hoc tests showed that thehighest 19.4 μg/kg dose decreased correct response latenciescompared to the vehicle treatment (p < 0.05). There was asignificant dose × group interaction on incorrect response la-tency (dose: F[3,87] = 0.309, p = 0.819; dose × group:F[3,87] = 3.356, p = 0.022), but post hoc tests revealed nosignificant pairwise comparisons.

Donepezil There was a strong but nonsignificant trend fordonepezil (0–1 mg/kg) to improve performance sensitivity, d′,with no impact on any other rCPT measures (Fig. 5a, b,Tables 2 and 3). One MAM-E17 and one sham rat had missingdata due to computer hardware failure during testing and wereexcluded from statistical analysis. There was a trend for a maineffect of donepezil to enhance d′ performance across bothMAM-E17 and sham control rats (dose: F[3,81] = 2.584,p = 0.064; dose × group: F[3,81] = 1.256, p < 0.295; H-Fcorrected). This trend toward performance enhancement wasmost pronounced at the lowest 0.1 mg/kg donepezil dose rela-tive to the vehicle treatment. There were no significant effectsof donepezil on any other performance measure in the rCPT.

EVP-6124 EVP-6124 (0–3 mg/kg) did not alter any perfor-mance measure in the rCPT (Fig. 5c, d, Tables 2 and 3). OneMAM-E17 animal made no responses at the 0.3 mg/kg doseand was thus necessarily excluded from statistical analyses ofd′, c, and response latencies. An identical pattern of statisticalresults was obtained if the animal was excluded from allmeasures.

There was a near-threshold significant main effect of EVP-6124 dose on correct response latency (dose: F[3,48] = 2.776,p = 0.050; dose × group: F[3,48] = 1.346, p = 0.270), wherethe largest effect was a slower correct response time at the1 mg/kg dose.


Discussion

MAM-E17 model impairments on rCPT

MAM-E17 rats showed robust and persistent deficits in dis-criminative sensitivity (d′) on the rCPT relative to sham con-trols. The effect size was large, virtually identical in two co-horts of animals (Cohen’s d = 1.23 and 1.27), and consistentacross repeated drug testing. This model profile is remarkablysimilar to impairments on common variants of human CPTswidely reported in neuropsychiatric disorders such as ADHDor schizophrenia (Chen and Faraone 2000; Riccio et al. 2002).For example, a meta-analysis examining CPT performanceacross 15 studies reported substantial d′ impairments(d = 1.18) in schizophrenia compared to control subjects(Heinrichs and Zakzanis 1998). A large, multi-site studyfound similar d′ decrements in schizophrenia on three-digit(d = 1.13) and four-digit (d = 1.14) identical-pairs CPT(Nuechterlein et al. 2015). Moreover, several studies havedemonstrated that such d′ deficits may remain stable over timeirrespective of psychotic/remitted state, positive or negativesymptoms or medication status, and are reliable across repeat-ed testing (Cornblatt et al. 1997; Nuechterlein et al. 2015).

The MAM-E17 deficits in d′ were observed in the absenceof significant changes in response bias/criterion (c). This dis-sociation between d′ and cmirrors what is commonly reportedin schizophrenia research using CPT (Mussgay and Hertwig1990; Nuechterlein et al. 2015). Response bias/criterion istheoretically independent of d′ and defines the extent to whicha particular response type or strategy (e.g., to respond or with-hold responding) is preferred or more probable (Stanislaw andTodorov 1999; Macmillan and Creelman 2004). In line withinterpretations of the human CPT literature, the robust andselective reduction in d′ for the MAM-E17 model may beattributable to impairments in sustained, focused, and/or se-lective attentional processes—the ability to maintain readinessand to detect, discriminate, and respond appropriately to rap-idly and successively presented target stimuli across time(Parasuraman 1979; Davies and Parasuraman 1982).

Schizophrenia-related deficits in d′ are frequently reportedin conjunction with decreases in the number or rate of detectedtarget stimuli (Cornblatt et al. 1989; Nuechterlein et al. 2015).In the current study, MAM-E17 rats displayed a consistent,but nonsignificant, decrement in target HR relative to shamcontrols. The discrepancy in effect sizes for HR between ro-dent and human versions might reflect differences in cost/benefit structure. In the rCPT, there is a high incentive foranimals to maximize target hits (immediate food rewards)with modest punishment (delay) for occasional errant re-sponses. MAM-E17 rats might thus compensate for a lowerd′ by slightly adjusting their response strategy (e.g., higher

FAR) to maximize rewards (Lynn and Barrett 2014). By con-trast, human CPT variants typically offer no immediate incen-tives to encourage target hits (Locke and Braver 2008). Itwould be interesting to examine how systematic alterationsin the cost/benefit matrix of the rCPT or human CPTs alterthe performance of theMAM-E17model and individuals withschizophrenia, respectively.

The MAM-E17 model also has a profile consistent withdeficits in inhibitory control on the rCPT, showing significantincreases in nontarget FAR and in the number of screentouches in the absence of any stimuli (ISI touches). Generallocomotor hyperactivity of MAM-E17 animals has been re-ported in several studies (Le Pen et al. 2006; Hazane et al.2009; Ratajczak et al. 2015; O’Reilly et al. 2016; but seeFlagstad et al. 2004) and discussed as a possible confound inthe assessment of cognitive function (O’Reilly et al. 2016).However, it is unlikely that excessive, nonspecific motor ac-tivity of MAM-E17 rats is a large contributing factor to cog-nitive performance in the rCPT as we observed no significantgroup differences on HR, overall response bias/criterion, orany reaction time latencies. More specific impairments in in-hibitory control have been proposed based on performance inother behavioral paradigms: MAM-E17 rats showed an in-creased number of responses, decreased efficiency and num-ber of rewards earned on a differential reinforcement of low-rate (DRL-20) task, as well as a nonsignificant trend to makemore premature responses in the 5-CSRTT (Featherstone et al.2007). Moreover, the robust pattern of MAM-E17 deficits inreversal and extra-dimensional shift learning is suggestive ofgeneral impairments in the ability to flexibly inhibit responsesto prepotent or behaviorally salient stimuli (Moore et al. 2006;Featherstone et al. 2007; Gastambide et al. 2012). Increasedfalse alarm rates are also frequently reported in schizophrenicpatients (Wohlberg and Kornetsky 1973; Birkett et al. 2007;Nuechterlein et al. 2015) and are found to be correlated withnegative (Nuechterlein et al. 1986) or disorganized symptomdomains of the disorder (Vollema and Postma 2002).

The observed deficits in attentional and inhibitory responsecontrol on the rCPT are predicted based on the knownneurodevelopmental sequelae of the MAM-E17 model(Lodge and Grace 2007; Chen et al. 2014; Perez et al. 2014;Grace 2016; Penschuck et al. 2006; Lodge et al. 2009;Maćkowiak et al. 2014). However, despite its similarities toschizophrenia and robust performance impairments on therCPT, it is noteworthy that previous studies have not readilyobserved specific deficits in attention and executive functionwithin the MAM-E17 model. Impairments in prepulse inhibi-tion, latent inhibition, set shifting, and learning or perfor-mance on spatial tasks are well documented in MAM-E17animals (Flagstad et al. 2005; Moore et al. 2006;Featherstone et al. 2007; Gastambide et al. 2012; O’Reilly


et al. 2016), but owing to factors such as hyperactivity and/orother behavioral processes, their precise links to attentionaland/or executive control are unclear. Moreover, when theMAM-E17 model was evaluated on tasks explicitly designedto be analogous to human CPTs such as the 5-CSRTTor SAT/dSAT, no significant differences from sham controls were ob-served on any performance measures (Featherstone et al.2007; Howe et al. 2015). The present rCPT was designed toincorporate key parameters that have been demonstrated to taxattentional and executive control processes in the human CPTparadigm. These parameters include the successive presenta-tion of stimuli which demands withholding responses on non-target trials, as well as the use of visual stimuli that requireboth detection and discrimination between target and nontar-gets matched for overall luminance (Parasuraman 1979; Seeet al. 1995). The 5-CSRTT, SAT/dSAT, and related paradigmsdo not require withholding responses on any trials and onlyrequire detection and/or localization of luminance increments.These task differences are likely to engage distinct neuralmechanisms and might help explain the increased sensitivityof the rCPT to detect impairments in attentional and executivecontrol in the MAM-E17 model. It should be noted that thepresent study examined stable rCPT performance in middle-aged, male rats to help make it comparable to previous studiesexamining attentional and executive performance in theMAM-E17 model. It would be worthwhile to extend investi-gations to include female subjects and to assess model differ-ences at an earlier developmental period—from late juvenileto young adulthood—which represents a critical time forschizophrenia onset and which may also provide a better po-tential window for successful treatment intervention.

Pharmacological effects of psychoactive compounds

Suppressant-like effects (sulpiride, atomoxetine,LSN2463359, RO4938581)

Acute administration of the D2/D3 receptor antagonist andneuroleptic, sulpiride (0–30mg/kg); the selective noradrenergicreuptake inhibitor, atomoxetine (0–1 mg/kg); the mGlu5RPAM, LSN2463359 (0–5 mg/kg); and the GABA(A)α5R par-tial inverse agonist, RO4938581 (0–10 mg/kg), each exertedsuppressant-like effects on rCPT performance by dose-dependently reducing indices of overall responding across allanimals. These reductions were evidenced by an increased re-sponse criterion, c, and decreased ISI touches, HR, and/or FAR.Given that incorrect and reward collection latencies were unal-tered by these compounds, the suppressant effects are unlikelyto be simply attributed to potentially confounding factors suchas motor depression or anhedonia. Moreover, with the excep-tion of the highest dose of atomoxetine, the general absence ofdrug effects on sensitivity (d′) further suggests that the effects

cannot be easily related to changes in sustained, focused atten-tion or visual discrimination processes.

The suppressant-like effects of these compounds are con-sistent with numerous prior studies and may suggest that theyare acting to reduce impulsive/motivated responding to salientcues predictive of reward. Numerous studies have describedD2/D3 antagonist-induced decreases in operant respondingfor reward (Harrison et al. 1997; Heath et al. 2015; Froger-Colléaux and Castagné 2016). Anti-impulsive effects ofatomoxetine have also been widely observed in both rodentsand humans tested on a variety of cognitive paradigms(Blondeau and Dellu-Hagedorn 2007; Navarra et al. 2008;Robinson et al. 2008; Paterson et al. 2011; Fernando et al.2011; Baarendse and Vanderschuren 2011; Robinson 2012;Chamberlain et al. 2006, 2007; Bari et al. 2009; Tomlinsonet al. 2014).The mGlu5R PAM, ADX47273, has been ob-served to reduce the number of impulsive responses in normalhealthy rats in the 5-CSRTT (Liu et al. 2008; Isherwood et al.2015), and RO4938581, as well as several mGlu5R PAMs, hasbeen demonstrated to reduce hyperactivity observed inmodelsof NMDAr antagonist or psychostimulant administration(Schlumberger et al. 2010; Bartolomé-Nebreda et al.2013; Conde-Ceide et al. 2016; Ballard et al. 2009;Redrobe et al. 2012).

None of these compounds exerted any beneficial effects onthe rCPT performance sensitivity index, d′. For sulpiride, thisis in line with the clinical literature indicating that currentneuroleptic treatment mechanisms do not significantly im-prove fundamental attentional impairments in schizophrenia(Cornblatt et al. 1997; Nuechterlein et al. 2015). However,atomoxetine has enhanced performance accuracy on certainsustained attention assays, particularly when long and/or var-iable ITIs have been introduced or manipulated (Jentsch et al.2008; Baarendse and Vanderschuren 2011; Robinson 2012;Tomlinson et al. 2014). Cognitive-enhancing effects ofmGlu5R PAMs and GABA(A)α5R receptor-negative alloste-ric modulators (α5NAMs) have been reported on other tasksassessing behavioral flexibility, aversive learning, recognitionmemory, and working memory in normal, healthy mice andrats (Darrah et al. 2008; Uslaner et al. 2009; Stefani andMoghaddam 2010; Fowler et al. 2011; Gilmour et al. 2013;Dawson et al. 2006; Atack et al. 2006). Moreover,LSN2463359 (2.5 mg/kg) was observed to selectively restoreimpairments in reversal learning and extra-dimensional setshifting selectively in MAM-E17 rats in a bowl-diggingparadigm (Gastambide et al. 2012). The putative anti-impulsive properties of these compounds may provide aparsimonious and testable explanation for the pattern ofnootropic effects on certain other behavioral assays.Enhancement of task performance by compounds that helpreduce impulsivity might be predicted within experimentalparadigms in which rapid stimulus processing andresponding are not required and where hyperactivity or


impulsivity in model animals may confound and/or impairperformance.

We observed detrimental effects of higher-dose (1.0 mg/kg) atomoxetine on d′. Such impairments may be predictedby inverted-U theories of catecholamine function (Aston-Jones and Cohen 2005; Robbins and Arnsten 2009) and havealso been observed in primates in a working memory task(Gamo et al. 2010) and a decision-making task in rodents(Silveira et al. 2016). The novel features of the rCPT relativeto other rodent attentional paradigms may render it more de-manding and more sensitive to detrimental effects of higherdoses of pro-catecholaminergic agents.

Sulpiride was the only drug to selectively alter rCPT per-formance of MAM-E17 animals by dose-dependently reduc-ing their elevated false alarm rate. This result fits well withfindings that sulpiride treatment can specifically remediatedysfunctional plasticity within hippocampal and frontostriatalcircuitry of MAM-E17 rats (Belujon et al. 2013; Belujon et al.2014). Moreover, antagonism of dopamine D2 and/or D3 re-ceptors in the nucleus accumbens core has been demonstratedto selectively improve inhibitory control in other model ani-mals (medial prefrontal cortex (mPFC) lesioned, amphet-amine injected, or selected high-impulsive animals) that ex-hibit high levels of impulsivity on the 5-CSRTT (Pattij et al.2006; Pezze et al. 2008; Besson et al. 2010). Systemic D2/D3antagonism may reduce motivational salience for stimuli andresponses governing rCPT performance (Treadway and Zald2013) and preferentially remediate an inhibitory control defi-cit in MAM-E17 animals that may result from aberrantfrontostriatal-based salience mechanisms, as hypothesizedfor schizophrenia (Kapur 2003).

Stimulant-like effects (modafinil, ABT-594)

The eugeroic agent, modafinil (8 mg/kg), was the only com-pound to cause significant improvements in sensitivity, d′, acommon index of sustained, focused attentional performancein the CPT paradigm, but did so only in control rats. Modafinilhas shown mixed effects for cognitive enhancements inhealthy human subjects but generally improves performanceon tests of sustained and focused attention at more difficulttask parameters using low to moderate doses (Scoriels et al.2013; Battleday and Brem 2015). While the majority of stud-ies using young, healthy rodents in the 5-CSRTT has foundeither no effect or reductions in attentional accuracy and/orinhibitory control (Milstein et al. 2003; Waters et al. 2005;Liu et al. 2010), such results may be explained by the use ofhigher dose ranges (>30 mg/kg) and/or insufficient task diffi-culty. We also found that higher doses of modafinil in therCPT increased FAR and ISI touches, decreased criterion (c),and a significant linear trend toward faster correct responselatencies across all animals. Together with our observed im-pairments in d′ at the 64 mg/kg dose, these results suggest that

the stimulant effects of modafinil at higher doses might besubject to a speed/accuracy tradeoff and be detrimental to taskperformance. Indeed, higher doses of modafinil (64–128 mg/kg) have been reported to induce locomotor hyperactivity inrats (Simon et al. 1995; McClellan and Spencer 1998). Similarto our rCPT results, low-dose modafinil (10 mg/kg) has beenobserved to improve performance on a stop signal reactiontime task specifically in low-performing rats, while impairingbehavior at higher 100 mg/kg doses (Eagle et al. 2007).Moreover, one study has reported an enhancement in atten-tional accuracy following repeated daily dosing of modafinil(8–64 mg/kg) in older rats having low baseline performancelevels on a three-stimulus detection task with variable stimu-lus durations and inter-trial intervals (Morgan et al. 2007).Thus, both the lower dose (8 mg/kg) and the increased diffi-culty of the rCPT over the 5-CSRTT (e.g., variable stimulusdurations and ITIs, plus discrimination versus simple detec-tion) may have rendered it more sensitive for detecting theattention-enhancing effects of modafinil.

It is unclear why modafinil enhanced sham performancebut was ineffective within the MAM-E17 model. One possi-bility is that disruptions of mPFC structure and function(Grace 2016), a brain region consistently implicated in suc-cessful CPT performance (Rosvold et al. 1956; Glosser andGoodglass 1990; Carter et al. 1998), might preclude the drugs’beneficial effects. Modafinil has been demonstrated to in-crease dopamine levels (de Saint Hilaire et al. 2001) and ac-tivity (Gozzi et al. 2012) within the mPFC of the rat and toalter PFC activity during cognitive demand in humans (Rasettiet al. 2010; Minzenberg et al. 2010). Interestingly, althoughmodafinil has been found to improve aspects of executivecontrol (Turner et al. 2004), it has generally not been foundto rectify sustained attentional deficits in schizophrenia (Sevyet al. 2005; Freudenreich et al. 2009; Kane et al. 2010).Further investigation into the mechanisms of modafinil onattention is warranted.

The selective α4β2nAChR agonist ABT-594 exertedstimulant-like effects on HR, FAR, c, ISI touches, and cor-rect response latency across both groups of animals on therCPT. Stimulant effects of α4β2 agonists have been ob-served in a variety of tasks including locomotor activity(Bannon et al. 1998) and the 5-CSRTT (Grottick andHiggins 2000; Hahn et al. 2003; Mohler et al. 2010). Wedid not see any ABT-594-related improvements in d′, whichsuggests that the drug is not augmenting the sustained ca-pacity to correctly discriminate target from nontarget stimuli.However, our observed increases in HR and rewards earnedand decreases in correct response latency are potential indi-ces of cognitive enhancement and consistent with effectsobserved in other attentional paradigms. Improvements inresponse accuracy, total correct, and speeded correct reactiontimes have also been observed under specific conditionswithin the 5-CSRTT and SAT/dSAT paradigms following


α4β2nAChR agonist treatment (McGaughy et al. 1999;Mohler et al. 2010; Howe et al. 2010). This pattern of resultsmight be due to α4β2 agonists acting to lower the thresholdfor go responses to visually detected stimuli (presence orabsence of light). This hypothesis would explain the selec-tive improvements in performance accuracy on detectiontasks only when baseline performance is low. Moreover, asd′ on the rCPT requires sustained attention for both visualdiscrimination and visual detection processes, it may be thatthe former is not improved by nicotine or α4β2nAChR ag-onists. This interpretation might also fit with human studiesshowing faster responding but inefficacy of nicotine againstthe sustained attention deficits induced by ketamine inhumans (D’Souza et al. 2012) and failure of theα4β2nAChR agonist AZD3480 to improve sustained atten-tion in schizophrenic patients when visual discriminationprocesses are required (Velligan et al. 2012).

Compounds with limited effects on rCPT performance(donepezil, EVP-6124)

We observed no significant effects of the cholinesterase inhib-itor, donepezil (0–1.0 mg/kg), on any measure of rCPT per-formance in MAM-E17 or sham controls. This result is con-sistent with recent reviews and large meta-analyses, suggest-ing that the impact of donepezil on sustained attentional per-formance in schizophrenia is equivocal (Singh et al. 2012;Thakurathi et al. 2013; Choi et al. 2013). There are also lim-ited effects of donepezil on attention in healthy individuals(Beglinger et al. 2004). Our data did, however, show a near-significant effect of donepezil dose on d′ across all animals(p = 0.064)—strongest at the lowest 0.1 mg/kg dose—but alsoshow mean improvements at 0.3 and 1.0 mg/kg relative tovehicle. Donepezil is known to enhance behavioral contrastsensitivity, particularly within sensitive spatial frequencyranges (Soma et al. 2013). Moreover, we have recently ob-served that donepezil can enhance d′ of DBAmice in the rCPTparadigm, specifically on more difficult, shorter-stimulus du-rations (Kim et al. 2015). The enhancing effects of donepezilon sustained and focused visual attention may thus dependcritically on the visual capacity of the subject and the visualproperties of the stimuli used. It would be worthwhile to ex-amine the effects of donepezil or other pro-cholinergic com-pounds both in the MAM-E17 model and in schizophreniapatients using a CPT-like paradigm in which task parameterssuch as contrast, spatial frequency, and stimulus duration areexpressly manipulated.

The partial α7nAChR agonist, EVP-6124 (0–3.0 mg/kg),did not affect rCPT performance in sham controls or MAM-E17 animals. There has been much clinical interest around thetherapeutic potential of α7nAChR stimulation (Garay et al.2016), and one recent study showed that EVP-6124 (1.0 mg/kg) improved attentional control selectively in a low-

performing subgroup of rats using the 5C-CPT (Haywardet al. 2017). However, acute α7nAChR administration gener-ally does not affect rodent attentional performance as measuredby the 5-CSRTT (Grottick et al. 2003; Hahn et al. 2003; Hoyleet al. 2006; Nilsson et al. 2016). Moreover, α7nAChR agonistshave repeatedly demonstrated nonsignificant effects onsustained attention in individuals with schizophrenia (Olincyet al. 2006; Freedman et al. 2008; Lieberman et al. 2013;Preskorn et al. 2014; Umbricht et al. 2014). EVP-6124 hasrecently failed a phase III trial with the endpoint of improvedcognitive function in schizophrenia due to a lack of efficacy(Fidler 2016). Thus, our lack of enhancing effects of EVP-6124on attention and executive function appears to be in generalagreement with the bulk of preclinical and clinical findings.

Summary and conclusions

MAM-E17 rats show a robust and reproducible pattern ofdeficits on a novel touchscreen continuous performance testthat closely mirrors impairments observed in schizophreniaand other disorders (Table 1). The specific deficits in sustainedattention/executive control found using the rCPT paradigmare predicted based on underlying neurobiological disruptionsin the MAM-E17 model yet are the first to be unambiguouslydetected within a rodent behavioral paradigm.

The eight acutely administered compounds tested in thisstudy each exhibited distinct behavioral profiles on rCPT per-formance variables as summarized in Table 2. Sulpiride,atomoxetine, LSN2463359, and RO493858 each showed sup-pressant effects on rCPT performance, including increases inresponse criterion (c) and decreases in ISI touches and HRand/or FAR. Conversely, modafinil and ABT-594 exertedstimulant-like effects with decreases in c and increases in ISItouches, HR, and/or FAR. Donepezil showed near-significantenhancements in d′, and EVP-6124 exerted no effects on therCPT. In terms of model-specific drug effects, sulpiride dose-dependently reduced the elevated FAR in MAM-E17 ratswhereas low-dose modafinil (8 mg/kg) improved d′ only insham controls. These drug profiles are largely consistent withthe human and rodent literatures using similar attentional par-adigms and highlight their observed general lack of effective-ness for selectively improving d′ in schizophrenia. The rCPTmay be an important translational test for the human CPTparadigm, which is one of the most widely used tools for theassessment of sustained and focused attention and executivefunction in clinical neuropsychometrics.

Acknowledgements The research leading to these results has receivedsupport from the Innovative Medicine Initiative Joint Undertaking undergrant agreement no. 115008 of which resources are composed of EFPIAin-kind contribution and financial contribution from the EuropeanUnion’s Seventh Framework Programme (FP7/2007-2013). TheBehavioural and Clinical Neuroscience Institute is co-funded by the


Medical Research Council and the Wellcome Trust. BGL receivedfunding support from the BLa Caixa^ Fellowship PostgraduateProgramme. ML was supported by the Chinese Scholarship CouncilJoint PhD Program. JA received funding support from the SwedishPharmaceutical Society and the Swedish Research Council (no. 350-2012-230). We would like to thank the following people for helpingprovide compounds used in the current studies: Dr. Eric Mohler,AbbVie (ABT-594); Dr. Niels Plath, H Lundbeck A/S (EVP-6924); Dr.Francois Gastambide, Eli Lilly and Company (Modafinil, LSN2463359);and Theresa Ballard, Roche (RO493858). We would also like to thankDhaarica Jeyanesan, Tazuko Edwards, Rebecca Woolley, and AndreaTrecker for their contributions to the behavioral testing of the first cohort.

Compliance with ethical standards All experiments were carried outin compliance with the UKAnimals (Scientific Procedures) Act 1986 andUK Animals (Scientific Procedures) Act 1986 Amendment Regulations2012.

Conflict of interest TWR discloses consultancy with CambridgeCognition, Lilly, Lundbeck, and GlaxoSmithKline and has researchgrants with Lilly, Lundbeck, and GlaxoSmithKline. LMS and TJB con-sult for Campden Instruments, Ltd. The other authors declare that theyhave no competing interests.

Open Access This article is distributed under the terms of the CreativeCommons At t r ibut ion 4 .0 In te rna t ional License (h t tp : / /creativecommons.org/licenses/by/4.0/), which permits unrestricted use,distribution, and reproduction in any medium, provided you giveappropriate credit to the original author(s) and the source, provide a linkto the Creative Commons license, and indicate if changes were made.

References

Abi-Dargham A, Rodenhiser J, Printz D et al (2000) Increased baselineoccupancy of D2 receptors by dopamine in schizophrenia. Proc NatlAcad Sci U S A 97:8104–8109

Aston-Jones G, Cohen JD (2005) An integrative theory of locuscoeruleus-norepinephrine function: adaptive gain and optimal per-formance. Annu Rev Neurosci 28:403–450. doi:10.1146/annurev.neuro.28.061604.135709

Atack JR, Bayley PJ, Seabrook GR et al (2006) L-655,708 enhancescognition in rats but is not proconvulsant at a dose selective forα5-containing GABAA receptors. Neuropharmacology 51:1023–1029. doi:10.1016/j.neuropharm.2006.04.018

Baarendse PJJ, Vanderschuren LJMJ (2011) Dissociable effects of mono-amine reuptake inhibitors on distinct forms of impulsive behavior inrats. Psychopharmacology 219:313–326. doi:10.1007/s00213-011-2576-x

Ballard TM, Knoflach F, Prinssen E et al (2009) RO4938581, a novelcognitive enhancer acting at GABAA alpha5 subunit-containingreceptors. Psychopharmacology 202:207–223. doi:10.1007/s00213-008-1357-7

Bannon AW, Decker MW, Curzon P et al (1998) ABT-594 [(r)-5-(2-azetidinylmethoxy)-2-chloropyridine]: a novel, orally effectiveantinociceptive agent acting vianeuronal nicotinic acetylcholine re-ceptors: II. In vivo characterization. J Pharmacol Exp Ther 285:787–794. doi:10.1002/(ISSN)2052-1707

Bari A, Eagle DM, Mar AC et al (2009) Dissociable effects of noradren-aline, dopamine, and serotonin uptake blockade on stop task perfor-mance in rats. Psychopharmacology 205:273–283. doi:10.1007/s00213-009-1537-0

Barry RJ, Clarke AR, Hajós M et al (2009) Acute atomoxetine effects onthe EEG of children with attention-deficit/hyperactivity disorder.

Neuropharmacology 57:702–707. doi:10.1016/j.neuropharm.2009.08.003

Bartolomé-Nebreda JM, Conde-Ceide S, Delgado F et al (2013)Dihydrothiazolopyridone derivatives as a novel family of positiveallosteric modulators of the metabotropic glutamate 5 (mGlu5) re-ceptor. J Med Chem 56:7243–7259. doi:10.1021/jm400650w

Batt leday RM, Brem AK (2015) Modafini l for cogni t iveneuroenhancement in healthy non-sleep-deprived subjects: a sys-tematic review. Eur Neuropsychopharmacol 25:1865–1881. doi:10.1016/j.euroneuro.2015.07.028

Beglinger LJ, Gaydos BL, Kareken DA et al (2004) Neuropsychologicaltest performance in healthy volunteers before and after donepeziladministration. J Psychopharmacol (Oxf) 18:102–108. doi:10.1177/0269881104040248

Belujon P, Patton MH, Grace AA (2014) Role of the prefrontal cortex inaltered hippocampal-accumbens synaptic plasticity in a develop-mental animal model of schizophrenia. Cereb Cortex 24:968–977.doi:10.1093/cercor/bhs380

Belujon P, PattonMH, Grace AA (2013)Disruption of prefrontal cortical-hippocampal balance in a developmental model of schizophrenia:reversal by sulpiride. Int J Neuropsychopharmacol 16:507–512. doi:10.1017/S146114571200106X

Benes FM, Vincent SL, Marie A, Khan Y (1996) Up-regulation ofGABAA receptor binding on neurons of the prefrontal cortex inschizophrenic subjects. Neuroscience 75:1021–1031. doi:10.1016/0306-4522(96)00328-4

Besson M, Belin D, McNamara R et al (2010) Dissociable control ofimpulsivity in rats by dopamine D2/3 receptors in the core and shellsubregions of the nucleus accumbens. Neuropsychopharmacology35:560–569. doi:10.1038/npp.2009.162

Bilder RM, Goldman RS, Robinson D et al (2000) Neuropsychology offirst-episode schizophrenia: initial characterization and clinical corre-lates. Am J Psychiatry 157:549–559. doi:10.1176/appi.ajp.157.4.549

Birkett P, Sigmundsson T, Sharma T et al (2007) Reaction time andsustained attention in schizophrenia and its genetic predisposition.Schizophr Res 95:76–85. doi:10.1016/j.schres.2007.05.030

Blondeau C, Dellu-Hagedorn F (2007) Dimensional analysis of ADHDsubtypes in rats. Biol Psychiatry 61:1340–1350. doi:10.1016/j.biopsych.2006.06.030

Buchanan RW, Strauss ME, Breier A et al (1997) Attentional impair-ments in deficit and nondeficit forms of schizophrenia. Am JPsychiatry 154:363–370. doi:10.1176/ajp.154.3.363

Cardinal RN, Aitken MRF (2006) ANOVA for the behavioural sciencesresearcher. L. Erlbaum, Mahwah, p xvi, 448

Carter CS, Braver TS, Barch DM et al (1998) Anterior cingulate cortex,error detection, and the online monitoring of performance. Science280:747–749

Chamberlain SR, del Campo N, Dowson J et al (2007) Atomoxetineimproved response inhibition in adults with attention deficit/hyperactivity disorder. Biol Psychiatry 62:977–984. doi:10.1016/j.biopsych.2007.03.003

Chamberlain SR, Müller U, Blackwell AD et al (2006) Neurochemicalmodulation of response inhibition and probabilistic learning inhumans. Science 311:861–863. doi:10.1126/science.1121218

Chen L, Perez SM, Lodge DJ (2014) An augmented dopamine systemfunction is present prior to puberty in the methylazoxymethanolacetate rodent model of schizophrenia. Dev Neurobiol 74:907–917. doi:10.1002/dneu.22172

Chen WJ, Faraone SV (2000) Sustained attention deficits as markers ofgenetic susceptibility to schizophrenia. Am J Med Genet 97:52–57

Choi KH, Wykes T, Kurtz MM (2013) Adjunctive pharmacotherapy forcognitive deficits in schizophrenia: meta-analytical investigation ofefficacy. Brit J Psychiatry 203:172–178. doi:10.1192/bjp.bp.111.107359

Conde-Ceide S, Alcázar J, Alonso de Diego SA et al (2016) Preliminaryinvestigation of 6,7-dihydropyrazolo[1,5-a]pyrazin-4-one


http://dx.doi.org/10.1146/annurev.neuro.28.061604.135709

http://dx.doi.org/10.1146/annurev.neuro.28.061604.135709

http://dx.doi.org/10.1016/j.neuropharm.2006.04.018

http://dx.doi.org/10.1007/s00213-011-2576-x

http://dx.doi.org/10.1007/s00213-011-2576-x

http://dx.doi.org/10.1007/s00213-008-1357-7

http://dx.doi.org/10.1007/s00213-008-1357-7

http://dx.doi.org/10.1002/(ISSN)2052-1707

http://dx.doi.org/10.1007/s00213-009-1537-0

http://dx.doi.org/10.1007/s00213-009-1537-0



http://dx.doi.org/10.1021/jm400650w

http://dx.doi.org/10.1016/j.euroneuro.2015.07.028

http://dx.doi.org/10.1177/0269881104040248

http://dx.doi.org/10.1177/0269881104040248

http://dx.doi.org/10.1093/cercor/bhs380

http://dx.doi.org/10.1017/S146114571200106X

http://dx.doi.org/10.1016/0306-4522(96)00328-4

http://dx.doi.org/10.1016/0306-4522(96)00328-4

http://dx.doi.org/10.1038/npp.2009.162

http://dx.doi.org/10.1176/appi.ajp.157.4.549

http://dx.doi.org/10.1016/j.schres.2007.05.030

http://dx.doi.org/10.1016/j.biopsych.2006.06.030


http://dx.doi.org/10.1176/ajp.154.3.363



http://dx.doi.org/10.1126/science.1121218

http://dx.doi.org/10.1002/dneu.22172

http://dx.doi.org/10.1192/bjp.bp.111.107359

http://dx.doi.org/10.1192/bjp.bp.111.107359

derivatives as a novel series of mGlu5 receptor positive allostericmodulators with efficacy in preclinical models of schizophrenia.Bioorg Med Chem Lett 26:429–434. doi:10.1016/j.bmcl.2015.11.098

Cornblatt B, Obuchowski M, Schnur DB, O’Brien JD (1997) Attentionand clinical symptoms in schizophrenia. Psychiatr Q 68:343–359.doi:10.1023/A:1025495030997

Cornblatt BA, Lenzenweger MF, Erlenmeyer-Kimling L (1989) The con-tinuous performance test, identical pairs version: II. Contrasting at-tentional profiles in schizophrenic and depressed patients.Psychiatry Res 29:65–85. doi:10.1016/0165-1781(89)90188-1

D’Souza DC, Ahn K, Bhakta S et al (2012) Nicotine fails to attenuateketamine-induced cognitive deficits and negative and positivesymptoms in humans: implications for schizophrenia. BiolPsychiatry 72:785–794. doi:10.1016/j.biopsych.2012.05.009

Darrah JM, Stefani MR, Moghaddam B (2008) Interaction of N-methyl-D-aspartate and group 5metabotropic glutamate receptors on behav-ioral flexibility using a novel operant set-shift paradigm. BehavPharmacol 19:225–234. doi:10.1097/FBP.0b013e3282feb0ac

Davies DR, Parasuraman R (1982) The psychology of vigilance.Academic, London

Dawson GR, Maubach KA, Collinson N et al (2006) An inverse agonistselective for alpha5 subunit-containing GABAA receptors enhancescognition. J Pharmacol Exp Ther 316:1335–1345. doi:10.1124/jpet.105.092320

de Saint Hilaire Z, Orosco M, Rouch C et al (2001) Variations in extra-cellular monoamines in the prefrontal cortex and medial hypothala-mus after modafinil administration: a microdialysis study in rats.Neuroreport 12:3533–3537. doi:10.1097/00001756-200111160-00032

Deakin JF, Slater P, Simpson MD et al (1989) Frontal cortical and lefttemporal glutamatergic dysfunction in schizophrenia. J Neurochem52:1781–1786

Dean AC, Sevak RJ, Monterosso JR et al (2011) Acute modafinil effectson attention and inhibitory control in methamphetamine-dependenthumans. J Stud Alcohol Drugs 72:943–953. doi:10.15288/jsad.2011.72.943

Donnelly-Roberts DL, Puttfarcken PS, Kuntzweiler TA et al (1998) ABT-594 [(R)-5-(2-azetidinylmethoxy)-2-chloropyridine]: a novel, orallyeffective analgesic acting via neuronal nicotinic acetylcholine recep-tors: I. In vitro characterization. J Pharmacol Exp Ther 285:777–786

Eagle DM, Tufft MRA, Goodchild HL, Robbins TW (2007) Differentialeffects of modafinil and methylphenidate on stop-signal reactiontime task performance in the rat, and interactions with the dopaminereceptor antagonist cis-flupenthixol. Psychopharmacology 192:193–206. doi:10.1007/s00213-007-0701-7

Featherstone RE, Burton CL, Coppa-Hopman R et al (2009) Gestationaltreatment with methylazoxymethanol (MAM) that disruptshippocampal-dependent memory does not alter behavioural re-sponse to cocaine. Pharmacol Biochem Behav 93:382–390. doi:10.1016/j.pbb.2009.05.010

Featherstone RE, Rizos Z, Nobrega JN et al (2007) Gestationalmethylazoxymethanol acetate treatment impairs select cognitivefunctions: parallels to schizophrenia. Neuropsychopharmacology32:483–492. doi:10.1038/sj.npp.1301223

Fernando ABP, Economidou D, Theobald DE et al (2011) Modulation ofhigh impulsivity and attentional performance in rats by selectivedirect and indirect dopaminergic and noradrenergic receptor ago-nists. Psychopharmacology 219:341–352. doi:10.1007/s00213-011-2408-z

Fidler B (2016) Restructuring looms for forum as neuro drug fails keyclinical test. In: Xconomy. http://www.xconomy.com/boston/2016/03/24/restructuring-looms-for-forum-as-neuro-drug-fails-key-clinical-test/#. Accessed 21 Feb 2017

Flagstad P, Glenthøj BY, Didriksen M (2005) Cognitive deficits causedby late gestational disruption of neurogenesis in rats: a preclinical

model of schizophrenia. Neuropsychopharmacology 30:250–260.doi:10.1038/sj.npp.1300625

Flagstad P, Mørk A, Glenthøj BYet al (2004) Disruption of neurogenesison gestational day 17 in the rat causes behavioral changes relevant topositive and negative schizophrenia symptoms and altersamphetamine-induced dopamine release in nucleus accumbens.Neuropsychopharmacology 29:2052–2064. doi:10.1038/sj.npp.1300516

Foldi NS,White REC, Schaefer LA (2005) Detecting effects of donepezilon visual selective attention using signal detection parameters inAlzheimer’s disease. Int J Geriatr Psychiatry 20:485–488. doi:10.1002/gps.1319

Fowler SW, Ramsey AK, Walker JM et al (2011) Functional interactionof mGlu5 and NMDA receptors in aversive learning in rats.Neurobiol Learn Mem 95:73–79. doi:10.1016/j.nlm.2010.11.009

Freedman R, Olincy A, Buchanan RWet al (2008) Initial phase 2 trial of anicotinic agonist in schizophrenia. Am J Psychiatry 165:1040–1047.doi:10.1176/appi.ajp.2008.07071135

Freudenreich O, Henderson DC, Macklin EA et al (2009) Modafinil forclozapine-treated schizophrenia patients: a double-blind, placebo-controlled pilot trial. J Clin Psychiatry 70:1674–1680. doi:10.4088/JCP.08m04683

Froger-Colléaux C, Castagné V (2016) Effects of baclofen and racloprideon reinstatement of cocaine self-administration in the rat. Eur JPharmacol 777:147–155. doi:10.1016/j.ejphar.2016.03.008

Gamo NJ, Wang M, Arnsten AFT (2010) Methylphenidate andatomoxetine enhance prefrontal function through α2-adrenergicand dopamine D1 receptors. J Am Acad Child Psychiatry 49:1011–1023. doi:10.1016/j.jaac.2010.06.015

Garay RP, Citrome L, Samalin L et al (2016) Therapeutic improvementsexpected in the near future for schizophrenia and schizoaffectivedisorder: an appraisal of phase III clinical trials of schizophrenia-targeted therapies as found in US and EU clinical trial registries.Expert Opin Pharmacother 14656566(2016):1149164. doi:10.1517/14656566.2016.1149164

Gastambide F, Cotel M-C, Gilmour G et al (2012) Selective remediationof reversal learning deficits in the neurodevelopmental MAMmodelof schizophrenia by a novel mGlu5 positive allosteric modulator.Neuropsychopharmacology 37:1057–1066. doi:10.1038/npp.2011.298

Gastambide F, Taylor AM, Palmer C et al (2015) Alterations in spatialmemory and anxiety in the MAM E17 rat model of hippocampalpathology in schizophrenia. Psychopharmacology 232:4099–4112.doi:10.1007/s00213-014-3862-1

Gilmour G, Broad LM, Wafford KA et al (2013) In vitro characterisationof the novel positive allosteric modulators of the mGlu5 receptor,LSN2463359 and LSN2814617, and their effects on sleep architec-ture and operant responding in the rat. Neuropharmacology 64:224–239. doi:10.1016/j.neuropharm.2012.07.030

Glosser G, Goodglass H (1990) Disorders in executive control functionsamong aphasic and other brain-damaged patients. J Clin ExpNeuropsychol 12:485–501. doi:10.1080/01688639008400995

Gourevitch R, Rocher C, PenGL et al (2004)Workingmemory deficits inadult rats after prenatal disruption of neurogenesis. BehavPharmacol 15:287–292. doi:10.1097/01.fbp.0000135703.48799.71

Gozzi A, Colavito V, Seke Etet PF et al (2012) Modulation of fronto-cortical activity by modafinil: a functional imaging and fos study inthe rat. Neuropsychopharmacology 37:822–837. doi:10.1038/npp.2011.260

Grace AA (2016) Dysregulation of the dopamine system in the patho-physiology of schizophrenia and depression. Nat Rev Neurosci 17:524–532. doi:10.1038/nrn.2016.57

Grottick AJ, Haman M, Wyler R, Higgins GA (2003) Reversal of avigilance decrement in the aged rat by subtype-selective nicotinicligands. Neuropsychopharmacology 28:880–887. doi:10.1038/sj.npp.1300102


http://dx.doi.org/10.1016/j.bmcl.2015.11.098

http://dx.doi.org/10.1016/j.bmcl.2015.11.098

http://dx.doi.org/10.1023/A:1025495030997

http://dx.doi.org/10.1016/0165-1781(89)90188-1


http://dx.doi.org/10.1097/FBP.0b013e3282feb0ac

http://dx.doi.org/10.1124/jpet.105.092320


http://dx.doi.org/10.1097/00001756-200111160-00032

http://dx.doi.org/10.1097/00001756-200111160-00032

http://dx.doi.org/10.15288/jsad.2011.72.943

http://dx.doi.org/10.15288/jsad.2011.72.943

http://dx.doi.org/10.1007/s00213-007-0701-7

http://dx.doi.org/10.1016/j.pbb.2009.05.010

http://dx.doi.org/10.1038/sj.npp.1301223

http://dx.doi.org/10.1007/s00213-011-2408-z

http://dx.doi.org/10.1007/s00213-011-2408-z

http://www.xconomy.com/boston/2016/03/24/restructuring-looms-for-forum-as-neuro-drug-fails-key-clinical-test/






http://dx.doi.org/10.1002/gps.1319

http://dx.doi.org/10.1002/gps.1319

http://dx.doi.org/10.1016/j.nlm.2010.11.009

http://dx.doi.org/10.1176/appi.ajp.2008.07071135

http://dx.doi.org/10.4088/JCP.08m04683

http://dx.doi.org/10.4088/JCP.08m04683

http://dx.doi.org/10.1016/j.ejphar.2016.03.008

http://dx.doi.org/10.1016/j.jaac.2010.06.015

http://dx.doi.org/10.1517/14656566.2016.1149164

http://dx.doi.org/10.1517/14656566.2016.1149164



http://dx.doi.org/10.1007/s00213-014-3862-1


http://dx.doi.org/10.1080/01688639008400995

http://dx.doi.org/10.1097/01.fbp.0000135703.48799.71



http://dx.doi.org/10.1038/nrn.2016.57



Grottick AJ, Higgins GA (2000) Effect of subtype selective nicotiniccompounds on attention as assessed by the five-choice serial reac-tion time task. Behav Brain Res 117:197–208

Hahn B, Sharples CGV, Wonnacott S et al (2003) Attentional effects ofnicotinic agonists in rats. Neuropharmacology 44:1054–1067. doi:10.1016/S0028-3908(03)00099-6

Harrison AA, Everitt BJ, Robbins TW (1997) Central 5-HT depletionenhances impulsive responding without affecting the accuracy ofattentional performance: interactions with dopaminergic mecha-nisms. Psychopharmacology 133:329–342. doi:10.1007/s002130050410

Hayward A, Adamson L, Neill JC (2017) Partial agonism at the α7nicotinic acetylcholine receptor improves attention, impulsive actionand vigilance in low attentive rats. Eur Neuropsychopharmacol:1–11. doi:10.1016/j.euroneuro.2017.01.013

Hazane F, Krebs M-O, Jay TM, Le Pen G (2009) Behavioral perturba-tions after prenatal neurogenesis disturbance in female rat. NeurotoxRes 15:311–320. doi:10.1007/s12640-009-9035-z

Heath CJ, Bussey TJ, Saksida LM (2015) Motivational assessment ofmice using the touchscreen operant testing system: effects of dopa-minergic drugs. Psychopharmacology 232:4043–4057. doi:10.1007/s00213-015-4009-8

Heinrichs RW, Zakzanis KK (1998) Neurocognitive deficit in schizophre-nia: a quantitative review of the evidence. Neuropsychology 12:426–445. doi:10.1037/0894-4105.12.3.426

Howe WM, Ji J, Parikh V et al (2010) Enhancement of attentional per-formance by selective stimulation of alpha4beta2(*) nAChRs: un-derlying cholinergic mechanisms. Neuropsychopharmacology 35:1391–1401. doi:10.1038/npp.2010.9

Howe WM, Tierney PL, Young DA et al (2015) MAM (E17) rodentdevelopmental model of neuropsychiatric disease: disruptions inlearning and dysregulation of nucleus accumbens dopamine release,but spared executive function. Psychopharmacology 232:4113–4127. doi:10.1007/s00213-015-3955-5

Hoyle E, Genn RF, Fernandes C, Stolerman IP (2006) Impaired perfor-mance of alpha7 nicotinic receptor knockout mice in the five-choiceserial reaction time task. Psychopharmacology 189:211–223. doi:10.1007/s00213-006-0549-2

Hradetzky E, Sanderson TM, Tsang TM et al (2012) Themethylazoxymethanol acetate (MAM-E17) rat model: molecular andfunctional effects in the hippocampus. Neuropsychopharmacology37:364–377. doi:10.1038/npp.2011.219

Hunter MD, Ganesan V, Wilkinson ID, Spence SA (2006) Impact ofmodafinil on prefrontal executive function in schizophrenia. Am JPsychiatry 163:2184–2186. doi:10.1176/appi.ajp.163.12.2184

Hutton SB, Puri BK, Duncan LJ et al (1998) Executive function in first-episode schizophrenia. Psychol Med 28:463–473

Isherwood SN, Pekcec A, Nicholson JR et al (2015) Dissociable effects ofmGluR5 allosteric modulation on distinct forms of impulsivity in rats:interaction with NMDA receptor antagonism. Psychopharmacology232:3327–3344. doi:10.1007/s00213-015-3984-0

Jentsch JD, Aarde SM, Seu E (2008) Effects of atomoxetine and methyl-phenidate on performance of a lateralized reaction time task in rats.Psychopharmacology 202:497–504. doi:10.1007/s00213-008-1181-0

Kane JM, D’Souza DC, Patkar AA et al (2010) Armodafinil as adjunctivetherapy in adults with cognitive deficits associated with schizophre-nia: a 4-week, double-blind, placebo-controlled study. J ClinPsychiatry 71:1475–1481. doi:10.4088/JCP.09m05950gry

Kapur S (2003) Psychosis as a state of aberrant salience: a frameworklinking biology, phenomenology, and pharmacology in schizophre-nia. Am J Psychiatry 160:13–23. doi:10.1176/appi.ajp.160.1.13

Keefe RSE, Bilder RM, Harvey PD et al (2006) Baseline neurocognitivedeficits in the CATIE schizophrenia trial. Neuropsychopharmacology31:2033–2046. doi:10.1038/sj.npp.1301072

Kim CH, Hvoslef-Eide M, Nilsson SRO et al (2015) The continuousperformance test (rCPT) for mice: a novel operant touchscreen test

of attentional function. Psychopharmacology 232:3947–3966. doi:10.1007/s00213-015-4081-0

Knowles EEM, David AS, Reichenberg A (2010) Processing speed def-icits in schizophrenia: reexamining the evidence. Am J Psychiatry167:828–835. doi:10.1176/appi.ajp.2010.09070937

Le Pen G, Gourevitch R, Hazane F et al (2006) Peri-pubertal maturationafter developmental disturbance: a model for psychosis onset in therat. Neuroscience 143:395–405. doi:10.1016/j.neuroscience.2006.08.004

Leeson VC, Robbins TW, Matheson E et al (2009) Discrimination learn-ing, reversal, and set-shifting in first-episode schizophrenia: stabilityover six years and specific associations with medication type anddisorganization syndrome. Biol Psychiatry 66:586–593. doi:10.1016/j.biopsych.2009.05.016

Lewis DA (2000)GABAergic local circuit neurons and prefrontal corticaldysfunction in schizophrenia. Brain Res Rev 31:270–276

Lieberman JA, Dunbar G, Segreti AC et al (2013) A randomized explor-atory trial of an α-7 nicotinic receptor agonist (TC-5619) for cogni-tive enhancement in schizophrenia. Neuropsychopharmacology 38:968–975. doi:10.1038/npp.2012.259

Liu F, Grauer S, Kelley C et al (2008) ADX47273 [S-(4-fluoro-phenyl)-{3-[3-(4-fluoro-phenyl)-[1,2,4]-oxadiazol-5-yl]-piperidin-1-yl}-methanone]: a novel metabotropic glutamate receptor 5-selectivepositive allosteric modulator with preclinical antipsychotic-likeand procognitive activities. J Pharmacol Exp Ther 327:827–839.doi:10.1124/jpet.108.136580

Liu Y-P, Tung C-S, Lin Y-L, Chuang C-H (2010) Wake-promoting agentmodafinil worsened attentional performance following REM sleepdeprivation in a young-adult rat model of 5-choice serial reactiontime task. Psychopharmacology 213:155–166. doi:10.1007/s00213-010-2019-0

Locke HS, Braver TS (2008) Motivational influences on cognitive con-trol: behavior, brain activation, and individual differences. CognAffect Behav Neurosci 8:99–112. doi:10.3758/CABN.8.1.99

Lodge DJ, Behrens MM, Grace AA (2009) A loss of parvalbumin-containing interneurons is associated with diminished oscillatoryactivity in an animal model of schizophrenia. J Neurosci 29:2344–2354. doi:10.1523/JNEUROSCI.5419-08.2009

Lodge DJ, Grace AA (2008) Hippocampal dysfunction and disruption ofdopamine system regulation in an animal model of schizophrenia.Neurotox Res 14:97–104. doi:10.1007/BF03033801

Lodge DJ, Grace AA (2007) Aberrant hippocampal activity underlies thedopamine dysregulation in an animal model of schizophrenia. JNeurosci 27:11424–11430. doi:10.1523/JNEUROSCI.2847-07.2007

Lynn SK, Barrett LF (2014) BUtilizing^ signal detection theory. PsycholSci 25:1663–1673. doi:10.1177/0956797614541991

Macmillan NA, Creelman CD (2004) Detection theory: a user’s guideMaćkowiak M, Bator E, Latusz J et al (2014) Prenatal MAM administra-

tion affects histone H3 methylation in postnatal life in the rat medialprefrontal cortex. Eur Neuropsychopharmacol 24:271–289. doi:10.1016/j.euroneuro.2013.05.013

Mar AC, Horner AE, Nilsson S et al (2013) The touchscreen operantplatform for assessing executive function in rats and mice. NatProtoc 8:1985–2005. doi:10.1038/nprot.2013.123

Matricon J, Bellon A, Frieling H et al (2010) Neuropathological andreelin deficiencies in the hippocampal formation of rats exposed tomam; differences and similarities with schizophrenia. PLoS One 5:e10291–e10217. doi:10.1371/journal.pone.0010291

Maziade M, Rouleau N, Lee B et al (2009) Atomoxetine and neuropsy-chological function in children with attention-deficit/hyperactivityd i so rde r : re su l t s o f a p i lo t s tudy. J Chi ld AdolescPsychopharmacol 19:709–718. doi:10.1089/cap.2008.0166

McClellan KJ, Spencer CM (1998) Modafinil: a review of its pharmacol-ogy and clinical efficacy in the management of narcolepsy. CNSDrugs 9:311–324. doi:10.2165/00023210-199809040-00006


http://dx.doi.org/10.1016/S0028-3908(03)00099-6

http://dx.doi.org/10.1007/s002130050410

http://dx.doi.org/10.1007/s002130050410


http://dx.doi.org/10.1007/s12640-009-9035-z

http://dx.doi.org/10.1007/s00213-015-4009-8

http://dx.doi.org/10.1007/s00213-015-4009-8

http://dx.doi.org/10.1037/0894-4105.12.3.426


http://dx.doi.org/10.1007/s00213-015-3955-5

http://dx.doi.org/10.1007/s00213-006-0549-2



http://dx.doi.org/10.1007/s00213-015-3984-0

http://dx.doi.org/10.1007/s00213-008-1181-0

http://dx.doi.org/10.4088/JCP.09m05950gry



http://dx.doi.org/10.1007/s00213-015-4081-0

http://dx.doi.org/10.1176/appi.ajp.2010.09070937

http://dx.doi.org/10.1016/j.neuroscience.2006.08.004

http://dx.doi.org/10.1016/j.neuroscience.2006.08.004





http://dx.doi.org/10.1007/s00213-010-2019-0

http://dx.doi.org/10.1007/s00213-010-2019-0

http://dx.doi.org/10.3758/CABN.8.1.99

http://dx.doi.org/10.1523/JNEUROSCI.5419-08.2009

http://dx.doi.org/10.1007/BF03033801

http://dx.doi.org/10.1523/JNEUROSCI.2847-07.2007

http://dx.doi.org/10.1177/0956797614541991



http://dx.doi.org/10.1038/nprot.2013.123

http://dx.doi.org/10.1371/journal.pone.0010291

http://dx.doi.org/10.1089/cap.2008.0166

http://dx.doi.org/10.2165/00023210-199809040-00006

McGaughy J, Decker MW, Sarter M (1999) Enhancement of sustainedattention performance by the nicotinic acetylcholine receptor agonistABT-418 in intact but not basal forebrain-lesioned rats.Psychopharmacology 144:175–182

Milstein JA, Dalley JW, Theobald DEH, Robbins TW (2003) Effect ofmodafinil on performance of the 5-choice serial reaction time task inrats. J Psychopharmacol G45

Minzenberg MJ, Yoon JH, Carter CS (2010) Modafinil modulation of thedefault mode network. Psychopharmacology 215:23–31. doi:10.1007/s00213-010-2111-5

Mohler EG, Franklin SR, Rueter LE et al (2010) ABT-594 improvesperformance in the 5-choice serial reaction time task under condi-tions of increased difficulty, sub-chronic dosing, and in poorly-performing subjects. Pharmacol Biochem Behav 95:146–157. doi:10.1016/j.pbb.2009.12.019

Moore H, Jentsch JD, Ghajarnia M et al (2006) A neurobehavioral sys-tems analysis of adult rats exposed to methylazoxymethanol acetateon e17: implications for the neuropathology of schizophrenia. BiolPsychiatry 60:253–264. doi:10.1016/j.biopsych.2006.01.003

Morgan RE, Crowley JM, Smith RH et al (2007) Modafinil improvesattention, inhibitory control, and reaction time in healthy, middle-aged rats. Pharmacol Biochem Behav 86:531–541. doi:10.1016/j.pbb.2007.01.015

Mussgay L, Hertwig R (1990) Signal detection indices in schizophrenicson a visual, auditory, and bimodal continuous performance test.Schizophr Res 3:303–310

Navarra R, Graf R, Huang Y et al (2008) Effects of atomoxetine andmethylphenidate on attention and impulsivity in the 5-choice serialreaction time test. Prog Neuropsychopharmacol Biol Psychiat 32:34–41. doi:10.1016/j.pnpbp.2007.06.017

Nilsson SRO, Celada P, Fejgin K et al (2016) A mouse model of the15q13.3 microdeletion syndrome shows prefrontal neurophysiolog-ical dysfunctions and attentional impairment. Psychopharmacology233:2151–2163. doi:10.1007/s00213-016-4265-2

Nuechterlein KH, Edell WS, Norris M, Dawson ME (1986) Attentionalvulnerability indicators, thought disorder, and negative symptoms.Schizophr Bull 12:408–426

Nuechterlein KH, GreenMF, CalkinsME et al (2015) Attention/vigilancein schizophrenia: performance results from a large multi-site studyof the consortium on the genetics of schizophrenia (COGS).Schizophr Res 163:38–46. doi:10.1016/j.schres.2015.01.017

Q’Reilly KC, Perica MI, Fenton AA (2016) Memory deficits with intactcognitive control in the methylazoxymethanol acetate (MAM) ex-posure model of neurodevelopmental insult. Neurobiol Learn Mem134 Pt B:294–303. doi:10.1016/j.nlm.2016.07.034

Olincy A, Harris JG, Johnson LL et al (2006) Proof-of-concept trial of analpha7 nicotinic agonist in schizophrenia. Arch Gen Psychiatry 63:630–638. doi:10.1001/archpsyc.63.6.630

Oomen CA, Hvoslef-EideM, Heath CJ, Mar AC, Horner AE, Bussey TJ,Saksida LM (2013) The touchscreen operant platform for testingworking memory and pattern separation in rats and mice. NatProtoc 8(10):2006–2021. doi:10.1038/nprot.2013.124

Parasuraman R (1979) Memory load and event rate control sensitivitydecrements in sustained attention. Science 205:924–927

Paterson NE, Ricciardi J, Wetzler C, Hanania T (2011) Sub-optimal per-formance in the 5-choice serial reaction time task in rats was sensi-tive to methylphenidate, atomoxetine and d-amphetamine, but unaf-fected by the COMT inhibitor tolcapone. Neurosci Res 69:41–50.doi:10.1016/j.neures.2010.10.001

Pattij T, JanssenMCW, Vanderschuren LJMJ et al (2006) Involvement ofdopamine D1 and D2 receptors in the nucleus accumbens core andshell in inhibitory response control. Psychopharmacology 191:587–598. doi:10.1007/s00213-006-0533-x

Penschuck S, Flagstad P, Didriksen M et al (2006) Decrease inparvalbumin-expressing neurons in the hippocampus and increasedphencycl id ine- induced locomotor act iv i ty in the ra t

methylazoxymethanol (MAM) model of schizophrenia. Eur JNeurosci 23:279–284. doi:10.1111/j.1460-9568.2005.04536.x

Perez SM, Chen L, Lodge DJ (2014) Alterations in dopamine systemfunction across the estrous cycle of the MAM rodent model ofschizophrenia. Psychoneuroendocrinology 47:88–97. doi:10.1016/j.psyneuen.2014.05.005

Pezze MA, Dalley JW, Robbins TW (2008) Remediation of attentionaldysfunction in rats with lesions of the medial prefrontal cortex byintra-accumbens administration of the dopamine D2/3 receptor an-tagonist sulpiride. Psychopharmacology 202:307–313. doi:10.1007/s00213-008-1384-4

Phillips KG, Cotel MC, McCarthy AP et al (2012) Differential effects ofNMDA antagonists on high frequency and gamma EEG oscillationsi n a neu rodeve lopmen t a l mode l o f s ch i zoph r en i a .Neuropharmacology 62:1359–1370. doi:10.1016/j.neuropharm.2011.04.006

Preskorn SH, Gawryl M, Dgetluck N et al (2014) Normalizing effects ofEVP-6124, an alpha-7 nicotinic partial agonist, on event-relatedpotentials and cognition. J Psychiatr Pract 20:12–24. doi:10.1097/01.pra.0000442935.15833.c5

Prickaerts J, van Goethem NP, Chesworth R et al (2012) EVP-6124, anovel and selective α7 nicotinic acetylcholine receptor partial ago-nist, improves memory performance by potentiating the acetylcho-line response of α7 nicotinic acetylcholine receptors.Neuropharmacology 62:1099–1110. doi:10.1016/j.neuropharm.2011.10.024

Randall DC, Fleck NL, Shneerson JM, File SE (2004) The cognitive-enhancing properties of modafinil are limited in non-sleep-deprived middle-aged volunteers. Pharmacol Biochem Behav 77:547–555. doi:10.1016/j.pbb.2003.12.016

Randall DC, Shneerson JM, File SE (2005) Cognitive effects ofmodafinilin student volunteers may depend on IQ. Pharmacol BiochemBehav82:133–139. doi:10.1016/j.pbb.2005.07.019

Rasetti R, Mattay VS, Stankevich B et al (2010) Modulatory effects ofmodafinil on neural circuits regulating emotion and cognition.Neuropsychopharmacology 35:2101–2109. doi:10.1038/npp.2010.83

Ratajczak P, Kus K, Murawiecka P et al (2015) Biochemical and cogni-tive impairments observed in animal models of schizophrenia in-duced by prenatal stress paradigm or methylazoxymethanol acetateadministration. Acta Neurobiol Exp 75:314–325

Redrobe JP, Elster L, Frederiksen K et al (2012) Negative modulation ofGABAA α5 receptors by RO4938581 attenuates discrete sub-chronic and early postnatal phencyclidine (PCP)-induced cognitivedeficits in rats. Psychopharmacology 221:451–468. doi:10.1007/s00213-011-2593-9

Riccio CA, Reynolds CR, Lowe P, Moore JJ (2002) The continuousperformance test: a window on the neural substrates for attention?Arch Clin Neuropsychol 17:235–272. doi:10.1016/S0887-6177(01)00111-1

Robbins TW, Arnsten AFT (2009) The neuropsychopharmacology offronto-executive function: monoaminergic modulation. Annu RevNeurosci 32:267–287. doi:10.1146/annurev.neuro.051508.135535

Robinson ESJ (2012) Blockade of noradrenaline re-uptake sites improvesaccuracy and impulse control in rats performing a five-choice serialreaction time tasks. Psychopharmacology 219:303–312. doi:10.1007/s00213-011-2420-3

Robinson ESJ, Eagle DM, Mar AC et al (2008) Similar effects of theselective noradrenaline reuptake inhibitor atomoxetine on three dis-tinct forms of impulsivity in the rat. Neuropsychopharmacology 33:1028–1037. doi:10.1038/sj.npp.1301487

Rosvold HE, Mirsky AF, Sarason I et al (1956) A continuous perfor-mance test of brain damage. J Consult Psychol 20:343–350. doi:10.1037/h0043220

Sahakian BJ, Coull JT (1993) Tetrahydroaminoacridine (THA) inAlzheimer’s disease: an assessment of attentional and mnemonic


http://dx.doi.org/10.1007/s00213-010-2111-5

http://dx.doi.org/10.1007/s00213-010-2111-5





http://dx.doi.org/10.1016/j.pnpbp.2007.06.017

http://dx.doi.org/10.1007/s00213-016-4265-2


http://dx.doi.org/10.1016/j.nlm.2016.07.034

http://dx.doi.org/10.1001/archpsyc.63.6.630

http://dx.doi.org/10.1038/nprot.2013.124

http://dx.doi.org/10.1016/j.neures.2010.10.001

http://dx.doi.org/10.1007/s00213-006-0533-x

http://dx.doi.org/10.1111/j.1460-9568.2005.04536.x

http://dx.doi.org/10.1016/j.psyneuen.2014.05.005

http://dx.doi.org/10.1016/j.psyneuen.2014.05.005

http://dx.doi.org/10.1007/s00213-008-1384-4

http://dx.doi.org/10.1007/s00213-008-1384-4



http://dx.doi.org/10.1097/01.pra.0000442935.15833.c5

http://dx.doi.org/10.1097/01.pra.0000442935.15833.c5






http://dx.doi.org/10.1007/s00213-011-2593-9

http://dx.doi.org/10.1007/s00213-011-2593-9

http://dx.doi.org/10.1016/S0887-6177(01)00111-1

http://dx.doi.org/10.1016/S0887-6177(01)00111-1

http://dx.doi.org/10.1146/annurev.neuro.051508.135535

http://dx.doi.org/10.1007/s00213-011-2420-3

http://dx.doi.org/10.1007/s00213-011-2420-3


http://dx.doi.org/10.1037/h0043220

function using CANTAB. Acta Neurol Scand 88:29–35. doi:10.1111/j.1600-0404.1993.tb04251.x

Schlumberger C, Pietraszek M, Gravius A, DanyszW (2010) Effects of apositive allosteric modulator of mGluR5ADX47273 on conditionedavoidance response and PCP-induced hyperlocomotion in the rat asmodels for schizophrenia. Pharmacol Biochem Behav 95:23–30.doi:10.1016/j.pbb.2009.12.002

Schmaal L, Joos L, Koeleman M et al (2013) Effects of modafinil onneural correlates of response inhibition in alcohol-dependent pa-tients. Biol Psychiatry 73:211–218. doi:10.1016/j.biopsych.2012.06.032

Scoriels L, Jones PB, Sahakian BJ (2013) Modafinil effects on cognitionand emotion in schizophrenia and its neurochemical modulation inthe brain. Neuropharmacology 64:168–184. doi:10.1016/j.neuropharm.2012.07.011

See JE, Howe SR, Warm JS, Dember WN (1995) Meta-analysis of thesensitivity decrement in vigilance. Psychol Bull 117:230–249. doi:10.1037/0033-2909.117.2.230

Sevy S, Rosenthal MH, Alvir J et al (2005) Double-blind, placebo-controlled study of modafinil for fatigue and cognition in schizo-phrenia patients treated with psychotropic medications. J ClinPsychiatry 66:839–843

Silveira MM, Murch WS, Clark L, Winstanley CA (2016) Chronicatomoxetine treatment during adolescence does not influencedecision-making on a rodent gambling task, but does modulate am-phetamine’s effect on impulsive action in adulthood. BehavPharmacol 27:350–363. doi:10.1097/FBP.0000000000000203

Simon P, Hémet C, RamassamyC, Costentin J (1995) Non-amphetaminicmechanism of stimulant locomotor effect of modafinil in mice. EurNeuropsychopharmacol 5:509–514

Singh J, Kour K, Jayaram M (2012) P17 acetylcholinesterase inhibitorsfor schizophrenia. J Neurol. doi:10.1136/jnnp-2012-303538.34

Snyder MA, Adelman AE, Gao W-J (2012) Gesta t ionalmethylazoxymethanol exposure leads to NMDAR dysfunction in hip-pocampus during early development and lasting deficits in learning.Neuropsychopharmacology 38:328–340. doi:10.1038/npp.2012.180

Soares BG, Fenton M, Chue P (2000) Sulpiride for schizophrenia.Cochrane Database Syst Rev:CD001162. doi:10.1002/14651858.CD001162

Soma S, Suematsu N, Shimegi S (2013) Cholinesterase inhibitor,donepezil, improves visual contrast detectability in freely behavingrats. Behav Brain Res 256:362–367. doi:10.1016/j.bbr.2013.08.022

Stanislaw H, Todorov N (1999) Calculation of signal detection theorymeasures. Behav Res Methods Instrum Comput 31:137–149

Stefani MR, Moghaddam B (2010) Activation of type 5 metabotropicglutamate receptors attenuates deficits in cognitive flexibility

induced by NMDA receptor blockade. Eur J Pharmacol 639:26–32. doi:10.1016/j.ejphar.2010.01.028

Thakurathi N, Vincenzi B, Henderson DC (2013) Assessing the prospectof donepezil in improving cognitive impairment in patients withschizophrenia. Expert Opin Investig Drugs 22:259–265. doi:10.1517/13543784.2013.750650

Tomlinson A, Grayson B, Marsh S et al (2014) Pay attention to impul-sivity: modelling low attentive and high impulsive subtypes of adultADHD in the 5-choice continuous performance task (5C-CPT) infemale rats. Eur Neuropsychopharmacol 24:1371–1380. doi:10.1016/j.euroneuro.2014.04.008

Treadway MT, Zald DH (2013) Parsing anhedonia: translational modelsof reward-processing deficits in psychopathology. Curr Dir PsycholSci 22:244–249. doi:10.1177/0963721412474460

Tsai G, Passani LA, Slusher BS et al (1995) Abnormal excitatory neuro-transmitter metabolism in schizophrenic brains. Arch GenPsychia t ry 52:829–836. doi :10.1001/archpsyc .1995.03950220039008

Turner DC, Clark L, Pomarol-Clotet E et al (2004) Modafinil improvescognition and attentional set shifting in patients with chronic schizo-phrenia. Neuropsychopharmacology 29:1363–1373. doi:10.1038/sj.npp.1300457

Umbricht D, Keefe RSE, Murray S et al (2014) A randomized, placebo-controlled study investigating the nicotinic α7 agonist, RG3487, forcognitive deficits in schizophrenia. Neuropsychopharmacology 39:1568–1577. doi:10.1038/npp.2014.17

Uslaner JM, Parmentier-Batteur S, Flick RB et al (2009) Dose-dependenteffect of CDPPB, the mGluR5 positive allosteric modulator, on rec-ognition memory is associated with GluR1 and CREB phosphoryla-tion in the prefrontal cortex and hippocampus. Neuropharmacology57:531–538. doi:10.1016/j.neuropharm.2009.07.022

Velligan D, Brenner R, Sicuro F et al (2012) Assessment of the effects ofAZD3480 on cognitive function in patients with schizophrenia.Schizophr Res 134:59–64. doi:10.1016/j.schres.2011.10.004

Vollema MG, Postma B (2002) Neurocognitive correlates of schizotypyin first degree relatives of schizophrenia patients. Schizophr Bull 28:367–377. doi:10.1093/oxfordjournals.schbul.a006946

Waters KA, Burnham KE, Q’Connor D et al (2005) Assessment ofmodafinil on attentional processes in a five-choice serial reactiontime test in the rat. J Psychopharmacol (Oxf) 19:149–158. doi:10.1177/0269881105048995

Wohlberg GW, Kornetsky C (1973) Sustained attention in remittedschizophrenics. Arch Gen Psychiatry 28:533–537

Wong DT, Threlkeld PG, Best KL, Bymaster FP (1982) A new inhibitorof norepinephrine uptake devoid of affinity for receptors in rat brain.J Pharmacol Exp Ther 222:61–65. doi:10.1002/(ISSN)2052-1707


http://dx.doi.org/10.1111/j.1600-0404.1993.tb04251.x

http://dx.doi.org/10.1111/j.1600-0404.1993.tb04251.x






http://dx.doi.org/10.1037/0033-2909.117.2.230

http://dx.doi.org/10.1097/FBP.0000000000000203

http://dx.doi.org/10.1136/jnnp-2012-303538.34


http://dx.doi.org/10.1002/14651858.CD001162

http://dx.doi.org/10.1002/14651858.CD001162

http://dx.doi.org/10.1016/j.bbr.2013.08.022

http://dx.doi.org/10.1016/j.ejphar.2010.01.028

http://dx.doi.org/10.1517/13543784.2013.750650

http://dx.doi.org/10.1517/13543784.2013.750650



http://dx.doi.org/10.1177/0963721412474460

http://dx.doi.org/10.1001/archpsyc.1995.03950220039008

http://dx.doi.org/10.1001/archpsyc.1995.03950220039008






http://dx.doi.org/10.1093/oxfordjournals.schbul.a006946

http://dx.doi.org/10.1177/0269881105048995

http://dx.doi.org/10.1177/0269881105048995

http://dx.doi.org/10.1002/(ISSN)2052-1707

mam-e17 rat model impairments on a novel continuous ... · original investigation mam-e17 rat model...

Documents